| Name | semantic-lexicon JSON |
| Version |
0.1.4
JSON |
| download |
| home_page | None |
| Summary | Neural semantic modeling toolkit with persona-aware generation and diagnostics. |
| upload_time | 2025-10-19 22:28:43 |
| maintainer | None |
| docs_url | None |
| author | Semantic Lexicon Team |
| requires_python | >=3.9 |
| license | Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
|
| keywords |
nlp
semantics
persona
knowledge-graph
cli
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# Semantic Lexicon
[](https://github.com/semantic-lexicon/Semantic-Lexicon/actions/workflows/cli-tests.yml)
Semantic Lexicon is a NumPy-first research toolkit that demonstrates persona-aware semantic modelling. The project packages a compact neural stack consisting of intent understanding, a light-weight knowledge network, persona management, and text generation into an automated Python library and CLI.
The name reflects the long-standing academic concept of the [semantic lexicon](https://en.wikipedia.org/wiki/Semantic_lexicon); this repository contributes an applied, open implementation that operationalises those ideas for persona-aware experimentation.
---
## Contents
- [Quick Start](#quick-start)
- [Citation](#citation)
- [Features](#features)
- [Installation](#installation)
- [Project Layout](#project-layout)
- [CLI Walkthrough](#cli-walkthrough)
- [Streams & Clipboard](#streams--clipboard)
- [TADKit](#tadkit--drop-in-logits-processor-and-cli)
- [PersonaRAG](#personarag--exp3-personas-with-decode-time-truth-gates)
- [Development Workflow](#development-workflow)
- [Contributing](#contributing)
- [Acknowledgments](#acknowledgments)
- [Author's Note](#authors-note)
- [Contact & Legal](#contact--legal)
---
## Quick Start
```bash
# 1. Install the CLI and core library
pip install "semantic-lexicon @ git+https://github.com/farukalpay/Semantic-Lexicon.git"
# 2. Materialise the sample workspace
semantic-lexicon prepare \
--intent src/semantic_lexicon/data/intent.jsonl \
--knowledge src/semantic_lexicon/data/knowledge.jsonl \
--workspace artifacts
# 3. Train and generate
semantic-lexicon train --workspace artifacts
semantic-lexicon generate "Outline transformer basics" --workspace artifacts
```
Sample output:
```text
Persona: generic
Response:
1. Transformers stack self-attention blocks so each token can weigh every other token without recurrent passes.
2. Positional encodings add order-aware vectors that let attention keep track of word positions.
3. Feed-forward layers plus residual and layer-norm steps stabilise optimisation and let depth scale cleanly.
```
- Prefer `semantic-lexicon generate -` to pipe prompts from other tools (`echo question | semantic-lexicon generate -`).
- `semantic-lexicon clipboard --workspace artifacts` mirrors the same workflow but seeds the prompt from your system clipboard.
## Citation
Semantic Lexicon operationalises the reproducible persona-aware pipeline introduced in the accompanying preprint. If you build on this toolkit, please cite the work so other researchers can trace the connection between the paper's methodology and this implementation.
```bibtex
@misc{alpay2025reproduciblescalablepipelinesynthesizing,
title={A Reproducible, Scalable Pipeline for Synthesizing Autoregressive Model Literature},
author={Faruk Alpay and Bugra Kilictas and Hamdi Alakkad},
year={2025},
eprint={2508.04612},
archivePrefix={arXiv},
primaryClass={cs.IR},
url={https://arxiv.org/abs/2508.04612},
}
```
You can read the preprint online at [https://arxiv.org/abs/2508.04612](https://arxiv.org/abs/2508.04612); it documents the scalable data curation and evaluation strategy that directly powers the automation, diagnostics, and persona controls exposed by this repository.
## Features
**Core modelling**
- Modular architecture spanning embeddings, intents, knowledge graphs, personas, and persona-aware generation.
- Deterministic NumPy training loops delivering reproducible optimisation.
- Graph-driven knowledge curation using SPPMI weighting, smoothed relevance, and greedy facility-location selection.
**Automation & tooling**
- Typer-powered CLI for preparation, training, diagnostics, and generation.
- Extensible configuration with dataclass-backed YAML/JSON loading.
- Diagnostics covering embeddings, intents, knowledge neighbours, personas, and generation previews.
- Documentation plus regression safeguards via MkDocs, pytest, prompt evaluations, ruff, mypy, and black.
**Decision-making & safety**
- EXP3 persona selection utilities for adversarial style experiments.
- Analytical guarantees through composite reward shaping, calibration, and regret tooling.
- Primal–dual safety tuning that balances exploration, pricing, and knowledge gates until residuals vanish.
## Installation
1. **Create a virtual environment** (recommended):
```bash
python -m venv .venv
source .venv/bin/activate
```
2. **Install Semantic Lexicon** using pip's `name @ URL` syntax (avoids future deprecation warnings while matching the hosted repository):
```bash
pip install "semantic-lexicon[dev,docs] @ git+https://github.com/farukalpay/Semantic-Lexicon.git"
```
### Optional extras
Use the table below to tailor optional extras to your workflow:
| Goal | Command |
| --- | --- |
| Minimal CLI & library | `pip install "semantic-lexicon @ git+https://github.com/farukalpay/Semantic-Lexicon.git"` |
| Docs + developer tooling | `pip install "semantic-lexicon[dev,docs] @ git+https://github.com/farukalpay/Semantic-Lexicon.git"` |
| TADKit demos (needs PyTorch + Streamlit) | `pip install "semantic-lexicon[tadkit] @ git+https://github.com/farukalpay/Semantic-Lexicon.git"` |
| PersonaRAG demos (LangChain/LangGraph stack) | `pip install "semantic-lexicon[personarag] @ git+https://github.com/farukalpay/Semantic-Lexicon.git"` |
> **Note:** `tadkit` relies on PyTorch for logits processing. Install a CPU build with
> `pip install torch --index-url https://download.pytorch.org/whl/cpu` if it is not already
> included in your environment.
## Project Layout
```
src/semantic_lexicon/
├── cli.py # Typer CLI entry point
├── config.py # Dataclass-backed configuration helpers
├── embeddings.py # GloVe-style embeddings and persistence
├── intent.py # NumPy multinomial logistic regression intents
├── knowledge.py # Simple relation network with gradient updates
├── persona.py # Persona profiles and blending logic
├── generator.py # Persona-aware response generation
├── model.py # Orchestration façade
├── training.py # Training pipeline and diagnostics integration
├── diagnostics.py # Structured diagnostics reports
├── utils/ # Tokenisation, seeding, and I/O helpers
└── data/ # Sample intent & knowledge datasets for tests
```
> **Tip:** `examples/` mirrors these modules with runnable notebooks and scripts so you can jump from the reference implementation to experiments quickly.
## CLI Walkthrough
Follow the full pipeline when you want to rebuild artefacts from raw data.
1. **Prepare the corpus** (optional if using bundled sample data):
```bash
semantic-lexicon prepare --intent src/semantic_lexicon/data/intent.jsonl --knowledge src/semantic_lexicon/data/knowledge.jsonl --workspace artifacts
```
2. **Train the model** (uses processed datasets in `artifacts/`):
```bash
semantic-lexicon train --workspace artifacts
```
The CLI saves embeddings, intent weights, and knowledge matrices to the workspace directory.
3. **Run diagnostics**:
```bash
semantic-lexicon diagnostics --workspace artifacts --output diagnostics.json
```
The command prints a JSON summary to stdout and optionally writes the report to disk.
```json
{
"embedding_stats": {"vocab_size": 512, "mean_norm": 8.42, "std_norm": 0.77},
"intent_accuracy": 0.94,
"knowledge_graph": {"edges": 2048, "avg_degree": 3.1},
"personas_evaluated": ["generic", "tutor", "coach"]
}
```
4. **Generate responses**:
```bash
semantic-lexicon generate "Explain neural networks" --workspace artifacts --persona tutor
```
Example response:
```text
Persona: tutor
Response:
- Neural networks stack layers of weighted sums and nonlinear activations that learn feature detectors automatically.
- Training adjusts the weights with backpropagation so the model minimises prediction error on labelled examples.
- Regularisation (dropout, weight decay) keeps the learned representations from overfitting and improves generalisation.
```
Optional CLI calls:
- Tight briefing
```bash
semantic-lexicon ask-tight "How can I improve my research talks?" --workspace artifacts --bullets 3
```
```text
• Outline the three-part story: problem framing, insight, and next steps.
• Rehearse with a timer so every visual lands within the planned beats.
• Script a final call-to-action that leaves the audience with one clear task.
```
- Inspect knowledge selection
```bash
semantic-lexicon knowledge "Summarise convolutional neural networks" --workspace artifacts
```
```json
{
"concepts": [
"convolutional kernels capture spatial structure",
"pooling layers balance invariance with detail",
"feature maps highlight class-specific patterns"
],
"relevance": 0.87,
"coverage": 0.74,
"diversity": 0.69
}
```
## Truth-Aware Decoding Walkthrough
Truth-aware decoding (TAD) combines a model's logits with declarative knowledge supplied by one or more safety oracles. Each decode step (a) queries the model for logits, (b) consults the oracle for a boolean allow/block mask plus diagnostic labels, (c) computes the probability mass that remains safe, and (d) either selects the highest-probability safe token or abstains when the safe mass falls below a configurable threshold. The [`semantic_lexicon.decoding_tad.truth_aware_decode`](src/semantic_lexicon/decoding_tad.py) loop is pure NumPy and logs every decision in a `TADStepLog`, making it easy to audit or integrate into research pipelines.
### Reproducing the toy capital-of demo
The repository includes a fully-worked, research-grade example that resolves the prompt *“Paris is the capital of …”* against a knowledge-base oracle. The script reproduces the toy model from the TAD unit tests, injects a fact table with a single triple `(Paris, capital_of, France)`, and records all decode-time telemetry to disk.
```bash
python examples/truth_aware_decode_demo.py
```
Running the demo prints the safe decoding trace and saves the structured log to `examples/logs/paris_capital_truth_aware_decode.json`:
```
Prompt tokens: <BOS> Paris is the capital of
Generated tokens: France <EOS>
Log written to examples/logs/paris_capital_truth_aware_decode.json
```
The JSON log captures every metric needed for a forensic audit. Each entry in `steps` reports the decode index `t`, the safe probability mass `pi_safe` (after masking), the selected token, the number of blocked vocabulary entries, and the oracle-provided reason labels. A shortened excerpt is shown below; the full file is part of the repository so you can cite or diff it in papers and lab notebooks.
```json
{
"prompt_tokens": ["<BOS>", "Paris", "is", "the", "capital", "of"],
"generated_tokens": ["France", "<EOS>"],
"abstained": false,
"steps": [
{
"t": 0,
"pi_safe": 0.3390092760113778,
"picked_token": "France",
"blocked_count": 3,
"reasons_for_picked": ["kb:required_object"]
},
{
"t": 1,
"pi_safe": 1.0,
"picked_token": "<EOS>",
"blocked_count": 0,
"reasons_for_picked": []
}
]
}
```
From here you can: (1) swap in the graph-backed oracle to ground against a larger knowledge base, (2) set `TADConfig.abstain_token` to emit a sentinel when `pi_safe` drops below the threshold, and (3) feed the logged `pi_safe` sequence into your own reliability analyses (e.g., cumulative risk bounds or safe-mass histograms). Because `truth_aware_decode` works with any `Oracle` implementation, PhD students can plug in bespoke symbolic checkers—factuality verifiers, contradiction detectors, or mathematical solvers—without touching the decoding loop itself.
## TADKit — drop-in logits processor and CLI
The repository now exposes a standalone [`tadkit`](src/tadkit) package so you can pip-install the truth-aware decoding utilities outside of the monolithic CLI. TADKit mirrors the “expected product” shown in the product brief:
- `TruthOracle` turns CSV/JSON/YAML rules into prompt-activated constraints.
- `TADLogitsProcessor` plugs into `transformers` generation loops and injects abstain tokens when a rule is violated.
- `TADTrace` logs token-level actions and renders console tables or Pandas dataframes for audits.
- `tadkit compile` converts spreadsheets to JSON payloads; `tadkit demo` spins up a tiny `sshleifer/tiny-gpt2` demo using the compiled oracle.
- `examples/tadkit_quickstart.py` and `examples/tadkit_streamlit_app.py` are copy-pasteable quickstarts, matching the walkthrough in the brief.
Install extras as needed:
```bash
pip install "git+https://github.com/farukalpay/Semantic-Lexicon.git#egg=semantic-lexicon[tadkit]"
tadkit compile capitals.csv --out oracle.json --tokenizer gpt2
tadkit demo --oracle oracle.json --model sshleifer/tiny-gpt2 \
--prompt "Q: What is the capital of France?\nA:"
```
> **Note:** `tadkit` relies on PyTorch for logits processing. Install a CPU
> build with `pip install torch --index-url https://download.pytorch.org/whl/cpu`
> if you do not already have `torch` available.
## PersonaRAG — EXP3 personas with decode-time truth gates
PersonaRAG is a thin layer on top of LangChain/LangGraph that routes tone, enforces truth, and records feedback telemetry. The [`personarag`](src/personarag) package exposes:
- `BrandStyle` persona descriptors.
- `PersonaPolicyEXP3` — contextual EXP3 with weight telemetry and bulk feedback helpers.
- `KnowledgeGate` — wraps LangChain LLMs and installs `TADLogitsProcessor` when the underlying model exposes Hugging Face hooks.
- `examples/personarag_quickstart.py` — the complete “expected product” script from the brief.
Install with optional dependencies when you want the full LangChain stack:
```bash
pip install "git+https://github.com/farukalpay/Semantic-Lexicon.git#egg=semantic-lexicon[personarag]"
python examples/personarag_quickstart.py
```
Decode-time gating is enabled automatically for Hugging Face models (local or
via LangChain wrappers). Hosted chat models (e.g., OpenAI) receive trace
metadata only.
`KnowledgeGate` attaches `trace.events` to `response_metadata` (when available), so observability dashboards can render trace heatmaps alongside persona win-rates and abstain telemetry.
## Knowledge Selection Playbook
The knowledge selector now treats every AGENTS.md instruction as a hard feasibility constraint. Broad concepts can still join the shortlist, but only when they collaborate with prompt-relevant anchors *and* all group bounds are respected.
> **Note:** The full mathematical specification for the selector — including the object definitions, scoring components, constraints, and optimisation guarantees — now lives in [`docs/articles/knowledge-selector.tex`](docs/articles/knowledge-selector.tex). The README keeps the practitioner-focused workflow and validation guidance below; consult the article whenever you need the derivations or precise notation.
### Workflow
1. **Graph construction.** Estimate shifted PPMI weights with smoothing \(p(i)^\gamma\); derive \(S\), \(D\), \(L\), and \(P\).
2. **Relevance smoothing.** Compute raw cosine relevance, solve the graph-regularised system, and classify on/off-topic nodes via the topic threshold.
3. **Anchoring.** Select anchors, compute personalised PageRank bridges, and form soft gates \(g_i\).
4. **Group configuration.** Register AGENTS.md groups with `set_concept_groups` and interval bounds with `set_group_bounds`; the selector automatically adds on/off-topic ratios.
5. **Greedy selection.** Evaluate admissible candidates, compute marginal coverage, cohesion, collaboration, and diversity, and add the best concept while updating group capacities.
6. **Reporting.** Emit the chosen concepts plus relevance, coverage, cohesion, collaboration, diversity, raw knowledge score, and mean gate.
Defaults \((\alpha, \lambda, \mu, \gamma, \tau, \lambda_1, \lambda_2, K, \tau_g, \text{on/off ratios}) = (0.12, 0.08, 0.5, 0.35, 0.1, 0.6, 0.4, 12, 0.08, 0.6/0.2/0.4)) ship in `KnowledgeConfig`. Additional per-group intervals can be supplied at runtime. The legacy phrase planner (MMR phrase selection with PMI bonuses) remains available inside the generator for reproducibility.
Use the CLI to inspect the concepts chosen for a prompt without rendering a full response:
```bash
semantic-lexicon knowledge "Explain matrix multiplication" --workspace artifacts
```
The JSON payload now includes gated relevance, coverage, cohesion, collaboration reward, log-det diversity, the raw knowledge score,
and the mean gate value across selected concepts.
### Go/No-Go Validation
Before shipping a new persona or pricing configuration, run the Go/No-Go suite to certify that knowledge selection obeys AGENTS.md, the deployment policy respects the exploration rules, and the off-policy lift is trustworthy.
1. **Rule feasibility.** Map each concept to its groups and bounds, count how many selections fall inside every group, and reject whenever any lower or upper bound is violated. `SelectionSpec` now bundles a `KnowledgeSignals` payload so the same object carries the calibrated knowledge metrics required later in the gate.
2. **Policy consistency.** For each logged step, rebuild the policy that was deployed using the stored logits, temperature, exploration mixture, and whichever penalty mode (prices or congestion) was active. The policy gate fails if any logged action falls below its exploration floor, if prices and congestion penalties are mixed, if the knowledge weight leaves the [0,1] range, or if the SNIPS floor dips below the exploration limit — guarding the AGENTS exploration guarantees.
3. **Off-policy value & fairness.** Using tuples (x_i, a_i, r_i, p_i) and the reconstructed target policy, compute SNIPS weights, the estimated value, and the effective sample size. Enforce a non-negative lower confidence bound on the lift, require the effective sample size to exceed one percent of the log length, and evaluate fairness either on action frequencies or KPI gaps via `FairnessConfig`.
4. **Price/congestion stability.** Aggregate the penalty vector each timestep and ensure the most recent window keeps total variation below the configured threshold. `StabilityCheckResult` records the peak deviation so you can tighten rho or beta when oscillations appear.
5. **Knowledge lift.** Compare the calibrated score and graph metrics captured in `KnowledgeSignals`. The gate demands the calibrated knowledge score stay above the trailing median and both coverage and cohesion deltas remain non-negative against the baseline selection size.
6. **Go/No-Go decision.** `run_go_no_go` wires the six checks together and emits a `GoNoGoResult` containing the selection feasibility, policy mode, OPE summary (with ESS target), stability diagnostics, and knowledge lift verdict. The `accepted` flag only flips to `True` when **every** gate passes. If any condition fails, follow the fix-once cascade in the specification — tweak the single knob (e.g., adjust `l_off`, `tau_g`, `eta`, or `rho`) and re-run the optimisation exactly once before re-testing.
### Primal–Dual Safety Gate Autotuning
Manual gate sweeps are still supported, but the preferred workflow is to run the projected primal–dual controller introduced in `semantic_lexicon.safety`. The controller now minimises the supplied objective while enforcing convex constraints, matching the textbook projected primal–dual loop.
```python
from semantic_lexicon.safety import (
ConstraintSpec,
GateBounds,
ObjectiveSpec,
run_primal_dual_autotune,
)
objective = ObjectiveSpec(
function=lambda params: params["x1"] ** 2
+ params["x2"] ** 2
- params["x1"]
- params["x2"],
gradient=lambda params: {
"x1": 2.0 * params["x1"] - 1.0,
"x2": 2.0 * params["x2"] - 1.0,
},
)
constraints = [
ConstraintSpec(
"linear",
lambda params: params["x1"] + params["x2"] - 1.0,
gradient=lambda params: {"x1": 1.0, "x2": 1.0},
)
]
result = run_primal_dual_autotune(
objective,
constraints,
initial_parameters={"x1": 0.2, "x2": 0.8},
parameter_names=("x1", "x2"),
bounds={
"x1": GateBounds(lower=0.0, upper=1.0),
"x2": GateBounds(lower=0.0, upper=1.0),
},
primal_step=0.2,
dual_step=0.4,
)
print("before", result.history[0])
print("after", result.parameters)
```
The first history entry captures the primal iterate after the initial step alongside its constraint violation, while the final snapshot records the tuned solution and dual multiplier. Swapping in exploration, fairness, or stability constraints follows the same pattern—only the callbacks change.
### Single-change presentation planner
When time only allows one tweak before a repeat talk, call `build_single_adjustment_plan()` to fetch a rehearsable experiment and a set of intent-hidden contingency moves. The helper keeps pacing and visuals frozen, picks *story beats* as the highest-leverage lever, and returns:
- A 20-minute rehearsal script that remaps the 12-minute slot into five beats, captures the headline you expect listeners to write down in each block, logs energy scores, and enforces a pass/fail line that demands fresh takeaways past minute seven.
- Five backup drills covering energy checkpoints, a slide trim for mixed audiences, a Q&A guardrail, a warmth-restoring micro-story, and a lighting plus breathing tweak for filler-word control.
```python
from semantic_lexicon.presentation import build_single_adjustment_plan
experiment, backups = build_single_adjustment_plan()
print(experiment.focus)
for move in backups:
print(move.label)
```
Backups remain intent-hidden so you can pivot mid-practice without exposing the heuristic to the audience.
## Lightweight Q&A Demo
Semantic Lexicon can answer short questions after its bundled model components are trained. The stack is intentionally tiny, so
the phrasing is concise, but the generator now runs a compact optimisation loop that:
1. **Classifies intent** with the logistic-regression intent model.
2. **Builds noun-phrase and collocation candidates** whose adjacent tokens clear an adaptive pointwise mutual information (PMI)
threshold, keeping multi-word ideas intact.
3. **Scores each candidate** via cosine relevance to the blended persona/prompt embedding, tf–idf salience, and a capped PMI
cohesion bonus.
4. **Selects diverse topics** with Maximum Marginal Relevance (MMR) plus an n-gram overlap penalty so the guidance does not echo
the question verbatim.
5. **Optimises knowledge coverage** by running the gated SPPMI graph objective (smoothed relevance, anchor gating, collaboration
reward, log-det diversity, and group-aware constraints) and appending the resulting knowledge focus and related concepts.
6. **Aligns journaling actions** with the detected intent so each topic carries a concise Explore/Practice/Reflect-style cue.
1. Install the project in editable mode:
```bash
pip install -e .
```
2. Run a quick script that trains the miniature model and generates answers for a few prompts:
```bash
python - <<'PY'
from semantic_lexicon import NeuralSemanticModel, SemanticModelConfig
from semantic_lexicon.training import Trainer, TrainerConfig
config = SemanticModelConfig()
model = NeuralSemanticModel(config)
trainer = Trainer(model, TrainerConfig())
trainer.train()
for prompt in [
"How do I improve my public speaking?",
"Explain matrix multiplication",
"What is machine learning?",
"Tips for staying productive while studying",
"Clarify the concept of photosynthesis",
"How can I organize my research presentation effectively?",
"Define gravitational potential energy",
]:
response = model.generate(prompt, persona="tutor")
print(
f"Prompt: {prompt}\\nResponse: {response.response}\\nKnowledge: {response.knowledge_hits}\\n"
)
PY
```
Sample output after training the bundled data:
```text
Prompt: How do I improve my public speaking?
Persona: tutor
Guidance:
- Schedule deliberate practice sessions (record short talks, review pacing and emphasis).
- Build a feedback loop with trusted listeners after each rehearsal.
- Reflect on audience energy so you can adjust tone and gesture.
Knowledge focus: practise short talks on camera.
Related concepts: collect feedback from trusted listeners; rehearse openings and transitions; track energy cues across slides.
Prompt: Explain matrix multiplication
Persona: tutor
Guidance:
- Describe matrix multiplication as repeated dot products between rows and columns.
- Connect the operation to linear transformations that reshape vectors.
- Compare 2×2 and 3×3 cases to build intuition about scaling and rotation.
Knowledge focus: review the row-by-column rule.
Related concepts: connect matrix products to linear transformations; practise multiplying 2×2 and 3×3 matrices; interpret column-space changes.
Prompt: Define gravitational potential energy
Persona: tutor
Guidance:
- State that gravitational potential energy equals mass × gravity × height relative to a reference.
- Show how choosing different reference frames shifts absolute values but not energy differences.
- Link the concept to conservation of mechanical energy in simple motion problems.
Knowledge focus: relate height changes to energy storage.
Related concepts: draw free-body diagrams for objects at different heights; compare gravitational and elastic potential energy; highlight conservation across motion phases.
```
These concise replies highlight the intentionally compact nature of the library's neural components—the toolkit is designed for
research experiments and diagnostics rather than fluent conversation, yet it showcases how questions can be routed through the
persona-aware pipeline.
Running `python examples/quickstart.py` (or `PYTHONPATH=src python examples/quickstart.py` from a checkout) produces a combined
generation preview and the new intent-selection walkthrough:
```
Sample generation:
Prompt: Share tips to learn python
Persona: tutor
Response: From a balanced tutor perspective, let's look at "Share tips to learn python." This ties closely to the "how_to" intent I detected. Consider journaling about: Study Schedule (Plan), Focus Blocks (Practice), Break Strategies (Reflect). Try to plan Study Schedule, practice Focus Blocks, and reflect on Break Strategies. Knowledge focus: schedule focused practice blocks. Related concepts worth exploring: work through bite-sized python projects, review core syntax and standard library patterns, reflect on debugging takeaways.
Journaling topics: Study Schedule, Focus Blocks, Break Strategies
Knowledge concepts: schedule focused practice blocks, work through bite-sized python projects, review core syntax and standard library patterns, reflect on debugging takeaways
Knowledge scores: relevance=3.956, coverage=0.865, cohesion=0.776, collaboration=0.349, diversity=6.867, K_raw=0.829, gate_mean=0.736
Calibration report: ECE raw=0.437 -> calibrated=0.027 (reduction=94%)
Reward weights: [0.2666619 0.2923091 0.075 0.366029 ]
Intent bandit walkthrough:
Prompt: Clarify when to use breadth-first search
Classifier intent: definition (optimal=definition)
Reward components: correctness=1.00, confidence=1.00, semantic=0.80, feedback=0.92
Composite reward: 0.96
Response: use case → shortest path in unweighted graphs; contrasts with → depth-first search
Prompt: How should I start researching renewable energy?
Classifier intent: how_to (optimal=how_to)
Reward components: correctness=1.00, confidence=0.45, semantic=0.80, feedback=0.92
Composite reward: 0.80
Response: first step → audit local energy use; research → read government energy outlook
Prompt: Compare supervised and unsupervised learning
Classifier intent: comparison (optimal=comparison)
Reward components: correctness=1.00, confidence=1.00, semantic=0.84, feedback=0.92
Composite reward: 0.96
Response: compare with → unsupervised learning; focus → labeled data; focus → pattern discovery
Prompt: Offer reflective prompts for creative writing
Classifier intent: exploration (optimal=exploration)
Reward components: correctness=1.00, confidence=0.42, semantic=0.80, feedback=0.92
Composite reward: 0.79
Response: prompt → explore character motivations; prompt → reflect on sensory details
```
The quickstart rewards are simulated using the intent classifier's posterior probabilities so the bandit loop stays in the unit
interval without external feedback.
You can opt into saving the calibrated accuracy curve and the empirical-vs-theoretical EXP3 regret comparison that back the
analysis appendix by setting `SEMANTIC_LEXICON_SAVE_PLOTS=1` (or `true/yes/on`) before running the script. This keeps the
repository free of bulky PNGs by default while still letting you regenerate them under `docs/assets/` on demand. Refer to the
generated CSV summaries in `Archive/` for the underlying values if you wish to recreate the plots with your preferred tooling.
The same behaviour is available through the CLI:
### Fixed-Point Ladders Companion
The research brief that motivated the README examples now has a full mathematical companion in
[`docs/articles/fixed-point-ladders.md`](docs/articles/fixed-point-ladders.md). The article walks through:
- **Parts A–C (Foundations & Logic):** proofs of the lattice background, the Knaster–Tarski theorem, Kleene iteration, and
µ-calculus semantics, all illustrated with the reachability operator that powers the persona-aware knowledge search.
- **Parts D–H (Shortcuts & Optimisation):** contraction-based accelerations, closure operators for finite-time stabilisation,
and multi-objective "best layer" selection rules that mirror the reward-shaping heuristics used in the quickstart bandit demo.
- **Parts I (Reflection):** a diagrammatic summary that ties the layer-by-layer iterations back to the automation loops in this
repository, making it easy to map the abstract ladders onto concrete CLI behaviours.
Each section keeps the ladder visual from the README and annotates it with the corresponding proofs or calculations so the
math-heavy readers can cross-check the guarantees while experimenting with the code.
For a dedicated, math-forward treatment of the fixed-point ladders referenced above, consult
[docs/articles/fixed-point-ladders.md](docs/articles/fixed-point-ladders.md).
```bash
semantic-lexicon generate "What is machine learning?" \
--workspace artifacts \
--persona tutor \
--config config.yaml
```
## Cross-domain validation & profiling
Run the bundled validation harness to stress-test the calibrated intent router on
100 prompts that span science, humanities, business, wellness, and personal
development queries:
```bash
PYTHONPATH=src python examples/cross_domain_validation.py
```
The script trains the classifier, evaluates it on the new prompt set, and saves a
report to `Archive/cross_domain_validation_report.json`. We report
\(\mathrm{Last}(\mathcal{R})\) and the corresponding content address \(h_{j^\star}\)
as defined in §4.
Runs are archived in `Archive/topic_pure_retrieval_runs.json`.
### §1 Core objects
```math
\begin{aligned}
&\mathbb{R}^d,\ d\in\mathbb{N}.\\
&\mathcal{C}=\{c_i\}_{i=1}^{N},\ E[c]\in\mathbb{R}^d.\\
&z:\mathcal{Q}\to \mathbb{R}^d.\\
&p\in\mathbb{R}^d\ \text{(use } p=0 \text{ if not applicable)}.\\
&g\in[0,1]^d,\quad M\succeq 0\in\mathbb{R}^{d\times d}.\\
&W=\Sigma^{-1/2}.
\end{aligned}
```
#### Scoring and retrieval
```math
r(q)=\mathrm{diag}(g)\,\big(z(q)+p\big),\qquad
s(q,c)=\big(Wr(q)\big)^{\!\top} M \big(WE[c]\big),\qquad
S_k(q)=\mathop{\mathrm{arg\,topk}}_{c\in\mathcal{C}} s(q,c).
```
### §2 Evaluation archive and identifiers
```math
R_j=\big(\Theta_j,\ \mathcal{D}_j,\ \mathcal{T}_j,\ \mathbf{m}_j,\ t_j\big).
```
```math
h_j=\mathsf{H}\!\big(\Theta_j,\,\mathcal{D}_j,\,\mathcal{T}_j\big),\quad
\text{with } \mathsf{H}:{\{0,1\}^\ast}\to\{0,1\}^{256} \text{ collision-resistant.}
```
### §3 Metrics
```math
\text{Purity@}k(q)=\frac{1}{k}\sum_{c\in S_k(q)}\mathbf{1}\{y(c)=y(q)\}.
```
```math
\mathsf{TVR}=\mathbb{P}\big[s(q,c^+)\le s(q,c^-)\big],\qquad
\mathsf{GS}=\frac{\|g\|_0}{d},\qquad
\kappa(\Sigma)=\frac{\lambda_{\max}(\Sigma)}{\lambda_{\min}(\Sigma)}.
```
```math
\mathbf{m}_j=\big(\ \overline{\text{Purity@}5},\ \overline{\text{Purity@}10},\ \mathsf{TVR},\ \mathsf{GS},\ \kappa(\Sigma)\ \big)_j,\quad
\overline{\cdot}\ \text{averages over } \mathcal{D}_j.
```
### §4 README functionals
```math
\mathrm{Last}(\mathcal{R})=\mathbf{m}_{j^\star},\quad j^\star=\arg\max_j t_j.
```
```math
\mathrm{Best}_{f,w}(\mathcal{R})=\mathbf{m}_{\arg\max_j f(w\odot \mathbf{m}_j)},\quad w\in\mathbb{R}_{\ge 0}^m,\ f:\mathbb{R}^m\to\mathbb{R}\ \text{monotone}.
```
```math
\mathrm{Mean}(\mathcal{R})=\frac{1}{n}\sum_{j=1}^{n}\mathbf{m}_j,\quad n=|\mathcal{R}|.
```
```math
\Delta(\mathcal{R})=\mathbf{m}_{j^\star}-\mathbf{m}_{j^\star-1}\quad (\text{defined if } n\ge 2).
```
```math
\big(h_{j^\star},\,\mathrm{Last}(\mathcal{R})\big)=\Big(\mathtt{845d7c3479535bdc83f7ed403e5b3695f242cc4561c807421f5c70d0c941291b},\ (0.6,0.5,0.0,1.0,371.6768300721485)\Big).
```
### §5 Example prompt I/O
```math
\mathcal{Q}^\star=\{q_0,q_1,q_2,q_3\},\qquad k=2.
```
```math
\Pi_k(q)=\big(S_k(q),\ s(q, S_k(q))\big).
```
```math
\text{Examples}(\mathcal{Q}^\star;\ h)=\Big\{\,\big(q,\ \Pi_k^{(h)}(q)\big)\ :\ q\in \mathcal{Q}^\star \Big\},\quad h=\mathtt{845d7c3479535bdc83f7ed403e5b3695f242cc4561c807421f5c70d0c941291b}.
```
### §6 Guarantees
- Lossless history: \(j\mapsto R_j\) is injective; the README exposes \(\{\mathbf{m}_j\}\) via \((h_{j^\star},\mathrm{Last}(\mathcal{R}))\).
- Determinism: for fixed \(h\) and \(q\), \(\Pi_k^{(h)}(q)\) is unique.
- Stability: \(P(P(M))=P(M),\ M\succeq 0\Rightarrow P(M)=M,\ P(M)\succeq 0\).
A companion benchmark is written to `Archive/intent_performance_profile.json`.
With heuristic fast paths, sparse dot products, and vector caching enabled the
optimised classifier processes repeated prompts **60 % faster** than the baseline
float64 pipeline (1.83 ms → 0.73 ms per request) while keeping the same accuracy.
Caching retains the most recent vectors, so the optimised pipeline uses ~27 KB of
RAM versus the baseline’s 4 KB; the additional footprint is documented alongside
the latency numbers so deployments can choose the appropriate trade-off.
## Streaming feedback API
Real-time user feedback can be folded into the composite reward with the new
HTTP server. Launch the background service by wiring an `IntentClassifier`
through `FeedbackService` and `FeedbackAPI`:
```python
from semantic_lexicon import IntentClassifier, IntentExample
from semantic_lexicon.api import FeedbackAPI, FeedbackService
from semantic_lexicon.utils import read_jsonl
examples = [
IntentExample(text=str(rec["text"]), intent=str(rec["intent"]), feedback=0.92)
for rec in read_jsonl("src/semantic_lexicon/data/intent.jsonl")
]
classifier = IntentClassifier()
classifier.fit(examples)
service = FeedbackService(classifier)
api = FeedbackAPI(service, host="127.0.0.1", port=8765)
api.start()
```
Submit streaming feedback with a simple POST request:
```bash
curl -X POST http://127.0.0.1:8765/feedback \
-H "Content-Type: application/json" \
-d '{"prompt": "Compare supervised and unsupervised learning", \
"selected_intent": "comparison", \
"optimal_intent": "comparison", \
"feedback": 0.96}'
```
The server replies with the updated composite-reward weights and the component
vector that was logged. Each event is processed under a lock so parallel clients
can stream feedback without clobbering the learned weights, and the new reward
weights remain simplex-projected for EXP3 compatibility.
Key parameters for `semantic-lexicon generate`:
- `--workspace PATH` – directory that contains the trained embeddings and weights (defaults to `artifacts`).
- `--persona NAME` – persona to blend into the response (defaults to the configuration's `default_persona`).
- `--config PATH` – optional configuration file to override model hyperparameters during loading.
## Adversarial Style Selection
Semantic Lexicon now bundles EXP3 helpers for experimenting with
adversarial persona *and* intent selection. The following snippet alternates
between two personas while learning from scalar feedback in ``[0, 1]``:
```python
from semantic_lexicon import AnytimeEXP3, NeuralSemanticModel, SemanticModelConfig
from semantic_lexicon.training import Trainer, TrainerConfig
config = SemanticModelConfig()
model = NeuralSemanticModel(config)
trainer = Trainer(model, TrainerConfig())
trainer.train()
bandit = AnytimeEXP3(num_arms=2)
personas = ["tutor", "researcher"]
for prompt in [
"Outline matrix factorisation for recommendations",
"Give journaling prompts about creativity",
"Explain reinforcement learning trade-offs",
]:
arm = bandit.select_arm()
persona = personas[arm]
response = model.generate(prompt, persona=persona)
score = min(1.0, len(response.response.split()) / 40.0)
bandit.update(score)
```
### Intent Selection with EXP3
We can model intent routing as an adversarial bandit problem. Let ``K`` be
the number of intents (e.g. ``{"how_to", "definition", "comparison", "exploration"}``).
At round ``t`` the system receives a prompt ``P_t`` and chooses an intent ``I_t``
using EXP3. After delivering the answer, a reward ``r_t`` in ``[0, 1]`` arrives
from explicit ratings or engagement metrics. The arm-selection probabilities are
$$
p_i(t) = (1 - \gamma) \frac{w_i(t)}{\sum_{j=1}^{K} w_j(t)} + \frac{\gamma}{K},
$$
and the weight for the played intent updates via
$$
w_{I_t}(t+1) = w_{I_t}(t) \exp\left(\frac{\gamma r_t}{K p_{I_t}(t)}\right).
$$
When the horizon ``T`` is unknown, the bundled ``AnytimeEXP3`` class applies the
doubling trick to refresh its parameters so the regret remains ``O(\sqrt{T})``.
The quickstart script demonstrates the pattern by mapping arms to intent labels
and simulating rewards from the classifier's posterior probability:
```python
from semantic_lexicon import AnytimeEXP3, NeuralSemanticModel, SemanticModelConfig
from semantic_lexicon.training import Trainer, TrainerConfig
config = SemanticModelConfig()
model = NeuralSemanticModel(config)
trainer = Trainer(model, TrainerConfig())
trainer.train()
intents = [label for _, label in sorted(model.intent_classifier.index_to_label.items())]
bandit = AnytimeEXP3(num_arms=len(intents))
prompt = "How should I start researching renewable energy?"
arm = bandit.select_arm()
intent = intents[arm]
reward = model.intent_classifier.predict_proba(prompt)[intent]
bandit.update(reward)
```
## Intent-Bandit Analysis Toolkit
The `semantic_lexicon.analysis` module supplies the maths underpinning the
improved EXP3 workflow:
- `RewardComponents` & `composite_reward` combine correctness, calibration,
semantic, and feedback signals into the bounded reward required by EXP3.
- `estimate_optimal_weights` fits component weights via simplex-constrained least
squares on historical interactions.
- `DirichletCalibrator` provides Bayesian confidence calibration with a
Dirichlet prior, yielding posterior predictive probabilities that minimise
expected calibration error.
- `simulate_intent_bandit` and `exp3_expected_regret` numerically check the
\(2.63\sqrt{K T \log K}\) regret guarantee for the composite reward.
- `compute_confusion_correction` and `confusion_correction_residual` extract the
SVD-based pseudoinverse that reduces systematic routing errors.
- `RobbinsMonroProcess` and `convergence_rate_bound` expose the stochastic
approximation perspective with an \(O(1/\sqrt{n})\) convergence rate bound.
See [docs/analysis.md](docs/analysis.md) for full derivations and proofs.
### Intent Classification Objective
Ethical deployment requires robust intent understanding. Semantic Lexicon's
``IntentClassifier`` treats intent prediction as a multinomial logistic regression
problem over prompts ``(P_i, I_i)``. Given parameters ``\theta``, the model
minimises the cross-entropy loss
$$
\mathcal{L}(\theta) = -\frac{1}{N} \sum_{i=1}^{N} \log p(I_i \mid P_i; \theta),
$$
which matches the negative log-likelihood optimised during training. Improving
intent accuracy directly translates into higher-quality feedback for the bandit
loop.
## Configuration
Semantic Lexicon reads configuration files in YAML or JSON using the `SemanticModelConfig` dataclass. Example `config.yaml`:
```yaml
embeddings:
dimension: 50
max_words: 5000
intent:
learning_rate: 0.2
epochs: 5
knowledge:
max_relations: 4
persona:
default_persona: tutor
generator:
temperature: 0.7
```
Load the configuration via CLI (`semantic-lexicon train --config config.yaml`) or programmatically:
```python
from semantic_lexicon import NeuralSemanticModel, load_config
config = load_config("config.yaml")
model = NeuralSemanticModel(config)
```
## Training API
```python
from semantic_lexicon import NeuralSemanticModel, SemanticModelConfig
from semantic_lexicon.training import Trainer, TrainerConfig
config = SemanticModelConfig()
model = NeuralSemanticModel(config)
trainer = Trainer(model, TrainerConfig())
trainer.train()
response = model.generate("How to learn python?", persona="tutor")
print(response.response)
```
## Diagnostics Programmatically
```python
from semantic_lexicon.model import NeuralSemanticModel
from semantic_lexicon.training import Trainer, TrainerConfig
model = NeuralSemanticModel()
trainer = Trainer(model, TrainerConfig())
trainer.train()
report = trainer.run_diagnostics()
print(report.to_dict())
```
## Development Workflow
| Task | Command |
| --------------- | --------------------------------- |
| Format & lint | `ruff check .` · `black .` |
| Type check | `mypy src` |
| Run tests | `pytest` |
| Preview docs | `mkdocs serve` |
A `Makefile` (or CI workflow) can orchestrate the tasks:
```bash
make lint
make test
make docs
```
## Streams & Clipboard
Generation now distinguishes abstract sources via the prompt functor \(𝐅\). Use cases:
- **Literal prompts** – pass a string and the CLI behaves exactly as before.
- **Streaming prompts** – pass `"-"` to fold STDIN chunks until EOF, perfect for shell pipelines.
- **Clipboard prompts** – call `semantic-lexicon clipboard` to pull the current system clipboard.
Example invocations:
```bash
echo "What is a transformer?" | semantic-lexicon generate - --workspace artifacts
semantic-lexicon clipboard --workspace artifacts --persona exploration
```
Both paths reuse the existing workspace/persona/config pipeline and reject empty inputs with a friendly error.
Sample outputs:
```text
$ echo "What is a transformer?" | semantic-lexicon generate - --workspace artifacts
Persona: generic
Response:
1. Transformers rely on self-attention so tokens draw context from the entire sentence in one step.
2. Multi-head attention lets the model track different relationships (syntax, long-range cues) simultaneously.
3. Decoder layers reuse the same mechanism to generate fluent text token by token.
$ semantic-lexicon clipboard --workspace artifacts --persona exploration
Clipboard prompt: "Give me three research angles on causal discovery."
Persona: exploration
Response:
1. Explore score-based causal discovery that leverages diffusion models to recover graph structure from noise.
2. Compare invariant risk minimisation versus meta-learning for handling interventions and domain shift.
3. Prototype active experimentation loops that query the system for the most informative interventions next.
```
---
## Contributing
1. Fork the repository and create a feature branch.
2. Install development dependencies: `pip install .[dev]`.
3. Run `make test` to ensure linting, typing, and tests pass.
4. Submit a pull request with detailed notes on new features or fixes.
## Acknowledgments
This work was shaped by the survey "Interpretation of Time-Series Deep Models: A Survey" [(arXiv:2305.14582)](https://arxiv.org/abs/2305.14582) shared by Dr. Zhao after reading our preprint on Calibrated "Counterfactual Conformal Fairness" (C3F) [(arXiv:2509.25295)](https://arxiv.org/abs/2509.25295). His survey offered both the conceptual framing and motivation for exploring this research path. We also thank Hamdi Alakkad and Bugra Kilictas for their pivotal contributions to our related preprints, which laid the groundwork for the developments presented here. We further acknowledge DeepSeek, whose advanced mathematical reasoning and logical inference capabilities substantially enhanced the precision and efficiency of the formal logic analysis, and the collaboration between OpenAI and GitHub on Codex, whose code generation strengths, in concert with DeepSeek’s systems, significantly accelerated and sharpened the overall development and analysis process.
## Author's Note
Hello people, or a system running perfectly, inbetween or broken -- At least working. -- While I am building groups, it is nice to see you behind them. This project represents my core self. We all came from a fixed point and would end up there as well. I am working on making myself “us,” me “our.” The physical world is for receiving and giving feelings, while the symbolic world is the projection of those feelings. Today is October 13, 2025, and I am located in Meckenheim, Germany. My plane landed yesterday from Istanbul—a nice trip, though (p.s. @farukalpayy). So, did you all feel like the energy was broken? It was the point where you get deep enough to realize where it was going. We reached the point where f(x) = x holds, but f(x) = y itself is also a point. And at this point, my request could be clarified. If this project saves you time or money, please consider sponsoring. Most importantly, it helps me keep improving and offering it free for the community. [Visit my Donation Page](https://buymeacoffee.com/farukalpay)
## Contact & Legal
- Semantic Lexicon is a Lightcap® research project distributed as open source under the Apache License 2.0; see [LICENSE](LICENSE) for details on rights and obligations.
- Lightcap® is a registered trademark (EUIPO Reg. No. 019172085).
- For enquiries, contact [alpay@lightcap.ai](mailto:alpay@lightcap.ai).
Raw data
{
"_id": null,
"home_page": null,
"name": "semantic-lexicon",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "nlp, semantics, persona, knowledge-graph, cli",
"author": "Semantic Lexicon Team",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/67/11/c53b3621ae21353320f462c670fdf809e6475c6c359aca96b99d0d83f645/semantic_lexicon-0.1.4.tar.gz",
"platform": null,
"description": "# Semantic Lexicon\n\n[](https://github.com/semantic-lexicon/Semantic-Lexicon/actions/workflows/cli-tests.yml)\n\nSemantic Lexicon is a NumPy-first research toolkit that demonstrates persona-aware semantic modelling. The project packages a compact neural stack consisting of intent understanding, a light-weight knowledge network, persona management, and text generation into an automated Python library and CLI.\n\nThe name reflects the long-standing academic concept of the [semantic lexicon](https://en.wikipedia.org/wiki/Semantic_lexicon); this repository contributes an applied, open implementation that operationalises those ideas for persona-aware experimentation.\n\n---\n\n## Contents\n\n- [Quick Start](#quick-start)\n- [Citation](#citation)\n- [Features](#features)\n- [Installation](#installation)\n- [Project Layout](#project-layout)\n- [CLI Walkthrough](#cli-walkthrough)\n- [Streams & Clipboard](#streams--clipboard)\n- [TADKit](#tadkit--drop-in-logits-processor-and-cli)\n- [PersonaRAG](#personarag--exp3-personas-with-decode-time-truth-gates)\n- [Development Workflow](#development-workflow)\n- [Contributing](#contributing)\n- [Acknowledgments](#acknowledgments)\n- [Author's Note](#authors-note)\n- [Contact & Legal](#contact--legal)\n\n---\n\n## Quick Start\n\n```bash\n# 1. Install the CLI and core library\npip install \"semantic-lexicon @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"\n\n# 2. Materialise the sample workspace\nsemantic-lexicon prepare \\\n --intent src/semantic_lexicon/data/intent.jsonl \\\n --knowledge src/semantic_lexicon/data/knowledge.jsonl \\\n --workspace artifacts\n\n# 3. Train and generate\nsemantic-lexicon train --workspace artifacts\nsemantic-lexicon generate \"Outline transformer basics\" --workspace artifacts\n```\n\nSample output:\n\n```text\nPersona: generic\nResponse:\n1. Transformers stack self-attention blocks so each token can weigh every other token without recurrent passes.\n2. Positional encodings add order-aware vectors that let attention keep track of word positions.\n3. Feed-forward layers plus residual and layer-norm steps stabilise optimisation and let depth scale cleanly.\n```\n\n- Prefer `semantic-lexicon generate -` to pipe prompts from other tools (`echo question | semantic-lexicon generate -`).\n- `semantic-lexicon clipboard --workspace artifacts` mirrors the same workflow but seeds the prompt from your system clipboard.\n\n## Citation\n\nSemantic Lexicon operationalises the reproducible persona-aware pipeline introduced in the accompanying preprint. If you build on this toolkit, please cite the work so other researchers can trace the connection between the paper's methodology and this implementation.\n\n```bibtex\n@misc{alpay2025reproduciblescalablepipelinesynthesizing,\n title={A Reproducible, Scalable Pipeline for Synthesizing Autoregressive Model Literature}, \n author={Faruk Alpay and Bugra Kilictas and Hamdi Alakkad},\n year={2025},\n eprint={2508.04612},\n archivePrefix={arXiv},\n primaryClass={cs.IR},\n url={https://arxiv.org/abs/2508.04612}, \n}\n```\n\nYou can read the preprint online at [https://arxiv.org/abs/2508.04612](https://arxiv.org/abs/2508.04612); it documents the scalable data curation and evaluation strategy that directly powers the automation, diagnostics, and persona controls exposed by this repository.\n\n## Features\n\n**Core modelling**\n- Modular architecture spanning embeddings, intents, knowledge graphs, personas, and persona-aware generation.\n- Deterministic NumPy training loops delivering reproducible optimisation.\n- Graph-driven knowledge curation using SPPMI weighting, smoothed relevance, and greedy facility-location selection.\n\n**Automation & tooling**\n- Typer-powered CLI for preparation, training, diagnostics, and generation.\n- Extensible configuration with dataclass-backed YAML/JSON loading.\n- Diagnostics covering embeddings, intents, knowledge neighbours, personas, and generation previews.\n- Documentation plus regression safeguards via MkDocs, pytest, prompt evaluations, ruff, mypy, and black.\n\n**Decision-making & safety**\n- EXP3 persona selection utilities for adversarial style experiments.\n- Analytical guarantees through composite reward shaping, calibration, and regret tooling.\n- Primal\u2013dual safety tuning that balances exploration, pricing, and knowledge gates until residuals vanish.\n\n## Installation\n\n1. **Create a virtual environment** (recommended):\n\n ```bash\n python -m venv .venv\n source .venv/bin/activate\n ```\n\n2. **Install Semantic Lexicon** using pip's `name @ URL` syntax (avoids future deprecation warnings while matching the hosted repository):\n\n ```bash\n pip install \"semantic-lexicon[dev,docs] @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"\n ```\n\n### Optional extras\n\nUse the table below to tailor optional extras to your workflow:\n\n| Goal | Command |\n| --- | --- |\n| Minimal CLI & library | `pip install \"semantic-lexicon @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"` |\n| Docs + developer tooling | `pip install \"semantic-lexicon[dev,docs] @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"` |\n| TADKit demos (needs PyTorch + Streamlit) | `pip install \"semantic-lexicon[tadkit] @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"` |\n| PersonaRAG demos (LangChain/LangGraph stack) | `pip install \"semantic-lexicon[personarag] @ git+https://github.com/farukalpay/Semantic-Lexicon.git\"` |\n\n> **Note:** `tadkit` relies on PyTorch for logits processing. Install a CPU build with\n> `pip install torch --index-url https://download.pytorch.org/whl/cpu` if it is not already\n> included in your environment.\n\n## Project Layout\n\n```\nsrc/semantic_lexicon/\n\u251c\u2500\u2500 cli.py # Typer CLI entry point\n\u251c\u2500\u2500 config.py # Dataclass-backed configuration helpers\n\u251c\u2500\u2500 embeddings.py # GloVe-style embeddings and persistence\n\u251c\u2500\u2500 intent.py # NumPy multinomial logistic regression intents\n\u251c\u2500\u2500 knowledge.py # Simple relation network with gradient updates\n\u251c\u2500\u2500 persona.py # Persona profiles and blending logic\n\u251c\u2500\u2500 generator.py # Persona-aware response generation\n\u251c\u2500\u2500 model.py # Orchestration fa\u00e7ade\n\u251c\u2500\u2500 training.py # Training pipeline and diagnostics integration\n\u251c\u2500\u2500 diagnostics.py # Structured diagnostics reports\n\u251c\u2500\u2500 utils/ # Tokenisation, seeding, and I/O helpers\n\u2514\u2500\u2500 data/ # Sample intent & knowledge datasets for tests\n```\n\n> **Tip:** `examples/` mirrors these modules with runnable notebooks and scripts so you can jump from the reference implementation to experiments quickly.\n\n## CLI Walkthrough\n\nFollow the full pipeline when you want to rebuild artefacts from raw data.\n\n1. **Prepare the corpus** (optional if using bundled sample data):\n\n ```bash\n semantic-lexicon prepare --intent src/semantic_lexicon/data/intent.jsonl --knowledge src/semantic_lexicon/data/knowledge.jsonl --workspace artifacts\n ```\n\n2. **Train the model** (uses processed datasets in `artifacts/`):\n\n ```bash\n semantic-lexicon train --workspace artifacts\n ```\n\nThe CLI saves embeddings, intent weights, and knowledge matrices to the workspace directory.\n\n3. **Run diagnostics**:\n\n ```bash\n semantic-lexicon diagnostics --workspace artifacts --output diagnostics.json\n ```\n\n The command prints a JSON summary to stdout and optionally writes the report to disk.\n\n ```json\n {\n \"embedding_stats\": {\"vocab_size\": 512, \"mean_norm\": 8.42, \"std_norm\": 0.77},\n \"intent_accuracy\": 0.94,\n \"knowledge_graph\": {\"edges\": 2048, \"avg_degree\": 3.1},\n \"personas_evaluated\": [\"generic\", \"tutor\", \"coach\"]\n }\n ```\n\n4. **Generate responses**:\n\n ```bash\n semantic-lexicon generate \"Explain neural networks\" --workspace artifacts --persona tutor\n ```\n\n Example response:\n\n ```text\n Persona: tutor\n Response:\n - Neural networks stack layers of weighted sums and nonlinear activations that learn feature detectors automatically.\n - Training adjusts the weights with backpropagation so the model minimises prediction error on labelled examples.\n - Regularisation (dropout, weight decay) keeps the learned representations from overfitting and improves generalisation.\n ```\n\nOptional CLI calls:\n\n- Tight briefing\n\n ```bash\n semantic-lexicon ask-tight \"How can I improve my research talks?\" --workspace artifacts --bullets 3\n ```\n\n ```text\n \u2022 Outline the three-part story: problem framing, insight, and next steps.\n \u2022 Rehearse with a timer so every visual lands within the planned beats.\n \u2022 Script a final call-to-action that leaves the audience with one clear task.\n ```\n\n- Inspect knowledge selection\n\n ```bash\n semantic-lexicon knowledge \"Summarise convolutional neural networks\" --workspace artifacts\n ```\n\n ```json\n {\n \"concepts\": [\n \"convolutional kernels capture spatial structure\",\n \"pooling layers balance invariance with detail\",\n \"feature maps highlight class-specific patterns\"\n ],\n \"relevance\": 0.87,\n \"coverage\": 0.74,\n \"diversity\": 0.69\n }\n ```\n\n## Truth-Aware Decoding Walkthrough\n\nTruth-aware decoding (TAD) combines a model's logits with declarative knowledge supplied by one or more safety oracles. Each decode step (a) queries the model for logits, (b) consults the oracle for a boolean allow/block mask plus diagnostic labels, (c) computes the probability mass that remains safe, and (d) either selects the highest-probability safe token or abstains when the safe mass falls below a configurable threshold. The [`semantic_lexicon.decoding_tad.truth_aware_decode`](src/semantic_lexicon/decoding_tad.py) loop is pure NumPy and logs every decision in a `TADStepLog`, making it easy to audit or integrate into research pipelines.\n\n### Reproducing the toy capital-of demo\n\nThe repository includes a fully-worked, research-grade example that resolves the prompt *\u201cParis is the capital of \u2026\u201d* against a knowledge-base oracle. The script reproduces the toy model from the TAD unit tests, injects a fact table with a single triple `(Paris, capital_of, France)`, and records all decode-time telemetry to disk.\n\n```bash\npython examples/truth_aware_decode_demo.py\n```\n\nRunning the demo prints the safe decoding trace and saves the structured log to `examples/logs/paris_capital_truth_aware_decode.json`:\n\n```\nPrompt tokens: <BOS> Paris is the capital of\nGenerated tokens: France <EOS>\nLog written to examples/logs/paris_capital_truth_aware_decode.json\n```\n\nThe JSON log captures every metric needed for a forensic audit. Each entry in `steps` reports the decode index `t`, the safe probability mass `pi_safe` (after masking), the selected token, the number of blocked vocabulary entries, and the oracle-provided reason labels. A shortened excerpt is shown below; the full file is part of the repository so you can cite or diff it in papers and lab notebooks.\n\n```json\n{\n \"prompt_tokens\": [\"<BOS>\", \"Paris\", \"is\", \"the\", \"capital\", \"of\"],\n \"generated_tokens\": [\"France\", \"<EOS>\"],\n \"abstained\": false,\n \"steps\": [\n {\n \"t\": 0,\n \"pi_safe\": 0.3390092760113778,\n \"picked_token\": \"France\",\n \"blocked_count\": 3,\n \"reasons_for_picked\": [\"kb:required_object\"]\n },\n {\n \"t\": 1,\n \"pi_safe\": 1.0,\n \"picked_token\": \"<EOS>\",\n \"blocked_count\": 0,\n \"reasons_for_picked\": []\n }\n ]\n}\n```\n\nFrom here you can: (1) swap in the graph-backed oracle to ground against a larger knowledge base, (2) set `TADConfig.abstain_token` to emit a sentinel when `pi_safe` drops below the threshold, and (3) feed the logged `pi_safe` sequence into your own reliability analyses (e.g., cumulative risk bounds or safe-mass histograms). Because `truth_aware_decode` works with any `Oracle` implementation, PhD students can plug in bespoke symbolic checkers\u2014factuality verifiers, contradiction detectors, or mathematical solvers\u2014without touching the decoding loop itself.\n\n## TADKit \u2014 drop-in logits processor and CLI\n\nThe repository now exposes a standalone [`tadkit`](src/tadkit) package so you can pip-install the truth-aware decoding utilities outside of the monolithic CLI. TADKit mirrors the \u201cexpected product\u201d shown in the product brief:\n\n- `TruthOracle` turns CSV/JSON/YAML rules into prompt-activated constraints.\n- `TADLogitsProcessor` plugs into `transformers` generation loops and injects abstain tokens when a rule is violated.\n- `TADTrace` logs token-level actions and renders console tables or Pandas dataframes for audits.\n- `tadkit compile` converts spreadsheets to JSON payloads; `tadkit demo` spins up a tiny `sshleifer/tiny-gpt2` demo using the compiled oracle.\n- `examples/tadkit_quickstart.py` and `examples/tadkit_streamlit_app.py` are copy-pasteable quickstarts, matching the walkthrough in the brief.\n\nInstall extras as needed:\n\n```bash\npip install \"git+https://github.com/farukalpay/Semantic-Lexicon.git#egg=semantic-lexicon[tadkit]\"\ntadkit compile capitals.csv --out oracle.json --tokenizer gpt2\ntadkit demo --oracle oracle.json --model sshleifer/tiny-gpt2 \\\n --prompt \"Q: What is the capital of France?\\nA:\"\n```\n\n> **Note:** `tadkit` relies on PyTorch for logits processing. Install a CPU\n> build with `pip install torch --index-url https://download.pytorch.org/whl/cpu`\n> if you do not already have `torch` available.\n\n## PersonaRAG \u2014 EXP3 personas with decode-time truth gates\n\nPersonaRAG is a thin layer on top of LangChain/LangGraph that routes tone, enforces truth, and records feedback telemetry. The [`personarag`](src/personarag) package exposes:\n\n- `BrandStyle` persona descriptors.\n- `PersonaPolicyEXP3` \u2014 contextual EXP3 with weight telemetry and bulk feedback helpers.\n- `KnowledgeGate` \u2014 wraps LangChain LLMs and installs `TADLogitsProcessor` when the underlying model exposes Hugging Face hooks.\n- `examples/personarag_quickstart.py` \u2014 the complete \u201cexpected product\u201d script from the brief.\n\nInstall with optional dependencies when you want the full LangChain stack:\n\n```bash\npip install \"git+https://github.com/farukalpay/Semantic-Lexicon.git#egg=semantic-lexicon[personarag]\"\npython examples/personarag_quickstart.py\n```\n\nDecode-time gating is enabled automatically for Hugging Face models (local or\nvia LangChain wrappers). Hosted chat models (e.g., OpenAI) receive trace\nmetadata only.\n\n`KnowledgeGate` attaches `trace.events` to `response_metadata` (when available), so observability dashboards can render trace heatmaps alongside persona win-rates and abstain telemetry.\n\n## Knowledge Selection Playbook\n\nThe knowledge selector now treats every AGENTS.md instruction as a hard feasibility constraint. Broad concepts can still join the shortlist, but only when they collaborate with prompt-relevant anchors *and* all group bounds are respected.\n\n> **Note:** The full mathematical specification for the selector \u2014 including the object definitions, scoring components, constraints, and optimisation guarantees \u2014 now lives in [`docs/articles/knowledge-selector.tex`](docs/articles/knowledge-selector.tex). The README keeps the practitioner-focused workflow and validation guidance below; consult the article whenever you need the derivations or precise notation.\n\n### Workflow\n\n1. **Graph construction.** Estimate shifted PPMI weights with smoothing \\(p(i)^\\gamma\\); derive \\(S\\), \\(D\\), \\(L\\), and \\(P\\).\n2. **Relevance smoothing.** Compute raw cosine relevance, solve the graph-regularised system, and classify on/off-topic nodes via the topic threshold.\n3. **Anchoring.** Select anchors, compute personalised PageRank bridges, and form soft gates \\(g_i\\).\n4. **Group configuration.** Register AGENTS.md groups with `set_concept_groups` and interval bounds with `set_group_bounds`; the selector automatically adds on/off-topic ratios.\n5. **Greedy selection.** Evaluate admissible candidates, compute marginal coverage, cohesion, collaboration, and diversity, and add the best concept while updating group capacities.\n6. **Reporting.** Emit the chosen concepts plus relevance, coverage, cohesion, collaboration, diversity, raw knowledge score, and mean gate.\n\nDefaults \\((\\alpha, \\lambda, \\mu, \\gamma, \\tau, \\lambda_1, \\lambda_2, K, \\tau_g, \\text{on/off ratios}) = (0.12, 0.08, 0.5, 0.35, 0.1, 0.6, 0.4, 12, 0.08, 0.6/0.2/0.4)) ship in `KnowledgeConfig`. Additional per-group intervals can be supplied at runtime. The legacy phrase planner (MMR phrase selection with PMI bonuses) remains available inside the generator for reproducibility.\nUse the CLI to inspect the concepts chosen for a prompt without rendering a full response:\n\n```bash\nsemantic-lexicon knowledge \"Explain matrix multiplication\" --workspace artifacts\n```\n\nThe JSON payload now includes gated relevance, coverage, cohesion, collaboration reward, log-det diversity, the raw knowledge score,\nand the mean gate value across selected concepts.\n\n### Go/No-Go Validation\n\nBefore shipping a new persona or pricing configuration, run the Go/No-Go suite to certify that knowledge selection obeys AGENTS.md, the deployment policy respects the exploration rules, and the off-policy lift is trustworthy.\n\n1. **Rule feasibility.** Map each concept to its groups and bounds, count how many selections fall inside every group, and reject whenever any lower or upper bound is violated. `SelectionSpec` now bundles a `KnowledgeSignals` payload so the same object carries the calibrated knowledge metrics required later in the gate.\n\n2. **Policy consistency.** For each logged step, rebuild the policy that was deployed using the stored logits, temperature, exploration mixture, and whichever penalty mode (prices or congestion) was active. The policy gate fails if any logged action falls below its exploration floor, if prices and congestion penalties are mixed, if the knowledge weight leaves the [0,1] range, or if the SNIPS floor dips below the exploration limit \u2014 guarding the AGENTS exploration guarantees.\n\n3. **Off-policy value & fairness.** Using tuples (x_i, a_i, r_i, p_i) and the reconstructed target policy, compute SNIPS weights, the estimated value, and the effective sample size. Enforce a non-negative lower confidence bound on the lift, require the effective sample size to exceed one percent of the log length, and evaluate fairness either on action frequencies or KPI gaps via `FairnessConfig`.\n\n4. **Price/congestion stability.** Aggregate the penalty vector each timestep and ensure the most recent window keeps total variation below the configured threshold. `StabilityCheckResult` records the peak deviation so you can tighten rho or beta when oscillations appear.\n\n5. **Knowledge lift.** Compare the calibrated score and graph metrics captured in `KnowledgeSignals`. The gate demands the calibrated knowledge score stay above the trailing median and both coverage and cohesion deltas remain non-negative against the baseline selection size.\n\n6. **Go/No-Go decision.** `run_go_no_go` wires the six checks together and emits a `GoNoGoResult` containing the selection feasibility, policy mode, OPE summary (with ESS target), stability diagnostics, and knowledge lift verdict. The `accepted` flag only flips to `True` when **every** gate passes. If any condition fails, follow the fix-once cascade in the specification \u2014 tweak the single knob (e.g., adjust `l_off`, `tau_g`, `eta`, or `rho`) and re-run the optimisation exactly once before re-testing.\n\n\n### Primal\u2013Dual Safety Gate Autotuning\n\nManual gate sweeps are still supported, but the preferred workflow is to run the projected primal\u2013dual controller introduced in `semantic_lexicon.safety`. The controller now minimises the supplied objective while enforcing convex constraints, matching the textbook projected primal\u2013dual loop.\n\n```python\nfrom semantic_lexicon.safety import (\n ConstraintSpec,\n GateBounds,\n ObjectiveSpec,\n run_primal_dual_autotune,\n)\n\nobjective = ObjectiveSpec(\n function=lambda params: params[\"x1\"] ** 2\n + params[\"x2\"] ** 2\n - params[\"x1\"]\n - params[\"x2\"],\n gradient=lambda params: {\n \"x1\": 2.0 * params[\"x1\"] - 1.0,\n \"x2\": 2.0 * params[\"x2\"] - 1.0,\n },\n)\n\nconstraints = [\n ConstraintSpec(\n \"linear\",\n lambda params: params[\"x1\"] + params[\"x2\"] - 1.0,\n gradient=lambda params: {\"x1\": 1.0, \"x2\": 1.0},\n )\n]\n\nresult = run_primal_dual_autotune(\n objective,\n constraints,\n initial_parameters={\"x1\": 0.2, \"x2\": 0.8},\n parameter_names=(\"x1\", \"x2\"),\n bounds={\n \"x1\": GateBounds(lower=0.0, upper=1.0),\n \"x2\": GateBounds(lower=0.0, upper=1.0),\n },\n primal_step=0.2,\n dual_step=0.4,\n)\n\nprint(\"before\", result.history[0])\nprint(\"after\", result.parameters)\n```\n\nThe first history entry captures the primal iterate after the initial step alongside its constraint violation, while the final snapshot records the tuned solution and dual multiplier. Swapping in exploration, fairness, or stability constraints follows the same pattern\u2014only the callbacks change.\n\n### Single-change presentation planner\n\nWhen time only allows one tweak before a repeat talk, call `build_single_adjustment_plan()` to fetch a rehearsable experiment and a set of intent-hidden contingency moves. The helper keeps pacing and visuals frozen, picks *story beats* as the highest-leverage lever, and returns:\n\n- A 20-minute rehearsal script that remaps the 12-minute slot into five beats, captures the headline you expect listeners to write down in each block, logs energy scores, and enforces a pass/fail line that demands fresh takeaways past minute seven.\n- Five backup drills covering energy checkpoints, a slide trim for mixed audiences, a Q&A guardrail, a warmth-restoring micro-story, and a lighting plus breathing tweak for filler-word control.\n\n```python\nfrom semantic_lexicon.presentation import build_single_adjustment_plan\n\nexperiment, backups = build_single_adjustment_plan()\nprint(experiment.focus)\nfor move in backups:\n print(move.label)\n```\n\nBackups remain intent-hidden so you can pivot mid-practice without exposing the heuristic to the audience.\n\n## Lightweight Q&A Demo\n\nSemantic Lexicon can answer short questions after its bundled model components are trained. The stack is intentionally tiny, so\nthe phrasing is concise, but the generator now runs a compact optimisation loop that:\n\n1. **Classifies intent** with the logistic-regression intent model.\n2. **Builds noun-phrase and collocation candidates** whose adjacent tokens clear an adaptive pointwise mutual information (PMI)\n threshold, keeping multi-word ideas intact.\n3. **Scores each candidate** via cosine relevance to the blended persona/prompt embedding, tf\u2013idf salience, and a capped PMI\n cohesion bonus.\n4. **Selects diverse topics** with Maximum Marginal Relevance (MMR) plus an n-gram overlap penalty so the guidance does not echo\n the question verbatim.\n5. **Optimises knowledge coverage** by running the gated SPPMI graph objective (smoothed relevance, anchor gating, collaboration\n reward, log-det diversity, and group-aware constraints) and appending the resulting knowledge focus and related concepts.\n6. **Aligns journaling actions** with the detected intent so each topic carries a concise Explore/Practice/Reflect-style cue.\n\n1. Install the project in editable mode:\n\n ```bash\n pip install -e .\n ```\n\n2. Run a quick script that trains the miniature model and generates answers for a few prompts:\n\n ```bash\n python - <<'PY'\n from semantic_lexicon import NeuralSemanticModel, SemanticModelConfig\n from semantic_lexicon.training import Trainer, TrainerConfig\n\n config = SemanticModelConfig()\n model = NeuralSemanticModel(config)\n trainer = Trainer(model, TrainerConfig())\n trainer.train()\n\n for prompt in [\n \"How do I improve my public speaking?\",\n \"Explain matrix multiplication\",\n \"What is machine learning?\",\n \"Tips for staying productive while studying\",\n \"Clarify the concept of photosynthesis\",\n \"How can I organize my research presentation effectively?\",\n \"Define gravitational potential energy\",\n ]:\n response = model.generate(prompt, persona=\"tutor\")\n print(\n f\"Prompt: {prompt}\\\\nResponse: {response.response}\\\\nKnowledge: {response.knowledge_hits}\\\\n\"\n )\n PY\n ```\n\n Sample output after training the bundled data:\n\n ```text\n Prompt: How do I improve my public speaking?\n Persona: tutor\n Guidance:\n - Schedule deliberate practice sessions (record short talks, review pacing and emphasis).\n - Build a feedback loop with trusted listeners after each rehearsal.\n - Reflect on audience energy so you can adjust tone and gesture.\n Knowledge focus: practise short talks on camera.\n Related concepts: collect feedback from trusted listeners; rehearse openings and transitions; track energy cues across slides.\n\n Prompt: Explain matrix multiplication\n Persona: tutor\n Guidance:\n - Describe matrix multiplication as repeated dot products between rows and columns.\n - Connect the operation to linear transformations that reshape vectors.\n - Compare 2\u00d72 and 3\u00d73 cases to build intuition about scaling and rotation.\n Knowledge focus: review the row-by-column rule.\n Related concepts: connect matrix products to linear transformations; practise multiplying 2\u00d72 and 3\u00d73 matrices; interpret column-space changes.\n\n Prompt: Define gravitational potential energy\n Persona: tutor\n Guidance:\n - State that gravitational potential energy equals mass \u00d7 gravity \u00d7 height relative to a reference.\n - Show how choosing different reference frames shifts absolute values but not energy differences.\n - Link the concept to conservation of mechanical energy in simple motion problems.\n Knowledge focus: relate height changes to energy storage.\n Related concepts: draw free-body diagrams for objects at different heights; compare gravitational and elastic potential energy; highlight conservation across motion phases.\n ```\n\n These concise replies highlight the intentionally compact nature of the library's neural components\u2014the toolkit is designed for\n research experiments and diagnostics rather than fluent conversation, yet it showcases how questions can be routed through the\n persona-aware pipeline.\n\n Running `python examples/quickstart.py` (or `PYTHONPATH=src python examples/quickstart.py` from a checkout) produces a combined\n generation preview and the new intent-selection walkthrough:\n\n ```\n Sample generation:\n Prompt: Share tips to learn python\n Persona: tutor\n Response: From a balanced tutor perspective, let's look at \"Share tips to learn python.\" This ties closely to the \"how_to\" intent I detected. Consider journaling about: Study Schedule (Plan), Focus Blocks (Practice), Break Strategies (Reflect). Try to plan Study Schedule, practice Focus Blocks, and reflect on Break Strategies. Knowledge focus: schedule focused practice blocks. Related concepts worth exploring: work through bite-sized python projects, review core syntax and standard library patterns, reflect on debugging takeaways.\n Journaling topics: Study Schedule, Focus Blocks, Break Strategies\n Knowledge concepts: schedule focused practice blocks, work through bite-sized python projects, review core syntax and standard library patterns, reflect on debugging takeaways\n Knowledge scores: relevance=3.956, coverage=0.865, cohesion=0.776, collaboration=0.349, diversity=6.867, K_raw=0.829, gate_mean=0.736\n\n Calibration report: ECE raw=0.437 -> calibrated=0.027 (reduction=94%)\n Reward weights: [0.2666619 0.2923091 0.075 0.366029 ]\n\n Intent bandit walkthrough:\n Prompt: Clarify when to use breadth-first search\n Classifier intent: definition (optimal=definition)\n Reward components: correctness=1.00, confidence=1.00, semantic=0.80, feedback=0.92\n Composite reward: 0.96\n Response: use case \u2192 shortest path in unweighted graphs; contrasts with \u2192 depth-first search\n\n Prompt: How should I start researching renewable energy?\n Classifier intent: how_to (optimal=how_to)\n Reward components: correctness=1.00, confidence=0.45, semantic=0.80, feedback=0.92\n Composite reward: 0.80\n Response: first step \u2192 audit local energy use; research \u2192 read government energy outlook\n\n Prompt: Compare supervised and unsupervised learning\n Classifier intent: comparison (optimal=comparison)\n Reward components: correctness=1.00, confidence=1.00, semantic=0.84, feedback=0.92\n Composite reward: 0.96\n Response: compare with \u2192 unsupervised learning; focus \u2192 labeled data; focus \u2192 pattern discovery\n\n Prompt: Offer reflective prompts for creative writing\n Classifier intent: exploration (optimal=exploration)\n Reward components: correctness=1.00, confidence=0.42, semantic=0.80, feedback=0.92\n Composite reward: 0.79\n Response: prompt \u2192 explore character motivations; prompt \u2192 reflect on sensory details\n ```\n\n The quickstart rewards are simulated using the intent classifier's posterior probabilities so the bandit loop stays in the unit\n interval without external feedback.\n\n You can opt into saving the calibrated accuracy curve and the empirical-vs-theoretical EXP3 regret comparison that back the\n analysis appendix by setting `SEMANTIC_LEXICON_SAVE_PLOTS=1` (or `true/yes/on`) before running the script. This keeps the\n repository free of bulky PNGs by default while still letting you regenerate them under `docs/assets/` on demand. Refer to the\n generated CSV summaries in `Archive/` for the underlying values if you wish to recreate the plots with your preferred tooling.\n The same behaviour is available through the CLI:\n\n### Fixed-Point Ladders Companion\n\nThe research brief that motivated the README examples now has a full mathematical companion in\n[`docs/articles/fixed-point-ladders.md`](docs/articles/fixed-point-ladders.md). The article walks through:\n\n- **Parts A\u2013C (Foundations & Logic):** proofs of the lattice background, the Knaster\u2013Tarski theorem, Kleene iteration, and\n \u00b5-calculus semantics, all illustrated with the reachability operator that powers the persona-aware knowledge search.\n- **Parts D\u2013H (Shortcuts & Optimisation):** contraction-based accelerations, closure operators for finite-time stabilisation,\n and multi-objective \"best layer\" selection rules that mirror the reward-shaping heuristics used in the quickstart bandit demo.\n- **Parts I (Reflection):** a diagrammatic summary that ties the layer-by-layer iterations back to the automation loops in this\n repository, making it easy to map the abstract ladders onto concrete CLI behaviours.\n\nEach section keeps the ladder visual from the README and annotates it with the corresponding proofs or calculations so the\nmath-heavy readers can cross-check the guarantees while experimenting with the code.\n\n For a dedicated, math-forward treatment of the fixed-point ladders referenced above, consult\n [docs/articles/fixed-point-ladders.md](docs/articles/fixed-point-ladders.md).\n\n```bash\nsemantic-lexicon generate \"What is machine learning?\" \\\n --workspace artifacts \\\n --persona tutor \\\n --config config.yaml\n```\n\n## Cross-domain validation & profiling\n\nRun the bundled validation harness to stress-test the calibrated intent router on\n100 prompts that span science, humanities, business, wellness, and personal\ndevelopment queries:\n\n```bash\nPYTHONPATH=src python examples/cross_domain_validation.py\n```\n\nThe script trains the classifier, evaluates it on the new prompt set, and saves a\nreport to `Archive/cross_domain_validation_report.json`. We report\n\\(\\mathrm{Last}(\\mathcal{R})\\) and the corresponding content address \\(h_{j^\\star}\\)\nas defined in \u00a74.\n\nRuns are archived in `Archive/topic_pure_retrieval_runs.json`.\n\n### \u00a71 Core objects\n\n```math\n\\begin{aligned}\n&\\mathbb{R}^d,\\ d\\in\\mathbb{N}.\\\\\n&\\mathcal{C}=\\{c_i\\}_{i=1}^{N},\\ E[c]\\in\\mathbb{R}^d.\\\\\n&z:\\mathcal{Q}\\to \\mathbb{R}^d.\\\\\n&p\\in\\mathbb{R}^d\\ \\text{(use } p=0 \\text{ if not applicable)}.\\\\\n&g\\in[0,1]^d,\\quad M\\succeq 0\\in\\mathbb{R}^{d\\times d}.\\\\\n&W=\\Sigma^{-1/2}.\n\\end{aligned}\n```\n\n#### Scoring and retrieval\n\n```math\nr(q)=\\mathrm{diag}(g)\\,\\big(z(q)+p\\big),\\qquad\ns(q,c)=\\big(Wr(q)\\big)^{\\!\\top} M \\big(WE[c]\\big),\\qquad\nS_k(q)=\\mathop{\\mathrm{arg\\,topk}}_{c\\in\\mathcal{C}} s(q,c).\n```\n\n### \u00a72 Evaluation archive and identifiers\n\n```math\nR_j=\\big(\\Theta_j,\\ \\mathcal{D}_j,\\ \\mathcal{T}_j,\\ \\mathbf{m}_j,\\ t_j\\big).\n```\n\n```math\nh_j=\\mathsf{H}\\!\\big(\\Theta_j,\\,\\mathcal{D}_j,\\,\\mathcal{T}_j\\big),\\quad\n\\text{with } \\mathsf{H}:{\\{0,1\\}^\\ast}\\to\\{0,1\\}^{256} \\text{ collision-resistant.}\n```\n\n### \u00a73 Metrics\n\n```math\n\\text{Purity@}k(q)=\\frac{1}{k}\\sum_{c\\in S_k(q)}\\mathbf{1}\\{y(c)=y(q)\\}.\n```\n\n```math\n\\mathsf{TVR}=\\mathbb{P}\\big[s(q,c^+)\\le s(q,c^-)\\big],\\qquad\n\\mathsf{GS}=\\frac{\\|g\\|_0}{d},\\qquad\n\\kappa(\\Sigma)=\\frac{\\lambda_{\\max}(\\Sigma)}{\\lambda_{\\min}(\\Sigma)}.\n```\n\n```math\n\\mathbf{m}_j=\\big(\\ \\overline{\\text{Purity@}5},\\ \\overline{\\text{Purity@}10},\\ \\mathsf{TVR},\\ \\mathsf{GS},\\ \\kappa(\\Sigma)\\ \\big)_j,\\quad\n\\overline{\\cdot}\\ \\text{averages over } \\mathcal{D}_j.\n```\n\n### \u00a74 README functionals\n\n```math\n\\mathrm{Last}(\\mathcal{R})=\\mathbf{m}_{j^\\star},\\quad j^\\star=\\arg\\max_j t_j.\n```\n\n```math\n\\mathrm{Best}_{f,w}(\\mathcal{R})=\\mathbf{m}_{\\arg\\max_j f(w\\odot \\mathbf{m}_j)},\\quad w\\in\\mathbb{R}_{\\ge 0}^m,\\ f:\\mathbb{R}^m\\to\\mathbb{R}\\ \\text{monotone}.\n```\n\n```math\n\\mathrm{Mean}(\\mathcal{R})=\\frac{1}{n}\\sum_{j=1}^{n}\\mathbf{m}_j,\\quad n=|\\mathcal{R}|.\n```\n\n```math\n\\Delta(\\mathcal{R})=\\mathbf{m}_{j^\\star}-\\mathbf{m}_{j^\\star-1}\\quad (\\text{defined if } n\\ge 2).\n```\n\n```math\n\\big(h_{j^\\star},\\,\\mathrm{Last}(\\mathcal{R})\\big)=\\Big(\\mathtt{845d7c3479535bdc83f7ed403e5b3695f242cc4561c807421f5c70d0c941291b},\\ (0.6,0.5,0.0,1.0,371.6768300721485)\\Big).\n```\n\n### \u00a75 Example prompt I/O\n\n```math\n\\mathcal{Q}^\\star=\\{q_0,q_1,q_2,q_3\\},\\qquad k=2.\n```\n\n```math\n\\Pi_k(q)=\\big(S_k(q),\\ s(q, S_k(q))\\big).\n```\n\n```math\n\\text{Examples}(\\mathcal{Q}^\\star;\\ h)=\\Big\\{\\,\\big(q,\\ \\Pi_k^{(h)}(q)\\big)\\ :\\ q\\in \\mathcal{Q}^\\star \\Big\\},\\quad h=\\mathtt{845d7c3479535bdc83f7ed403e5b3695f242cc4561c807421f5c70d0c941291b}.\n```\n\n### \u00a76 Guarantees\n\n- Lossless history: \\(j\\mapsto R_j\\) is injective; the README exposes \\(\\{\\mathbf{m}_j\\}\\) via \\((h_{j^\\star},\\mathrm{Last}(\\mathcal{R}))\\).\n- Determinism: for fixed \\(h\\) and \\(q\\), \\(\\Pi_k^{(h)}(q)\\) is unique.\n- Stability: \\(P(P(M))=P(M),\\ M\\succeq 0\\Rightarrow P(M)=M,\\ P(M)\\succeq 0\\).\n\nA companion benchmark is written to `Archive/intent_performance_profile.json`.\nWith heuristic fast paths, sparse dot products, and vector caching enabled the\noptimised classifier processes repeated prompts **60\u202f% faster** than the baseline\nfloat64 pipeline (1.83\u202fms \u2192 0.73\u202fms per request) while keeping the same accuracy.\nCaching retains the most recent vectors, so the optimised pipeline uses ~27\u202fKB of\nRAM versus the baseline\u2019s 4\u202fKB; the additional footprint is documented alongside\nthe latency numbers so deployments can choose the appropriate trade-off.\n\n## Streaming feedback API\n\nReal-time user feedback can be folded into the composite reward with the new\nHTTP server. Launch the background service by wiring an `IntentClassifier`\nthrough `FeedbackService` and `FeedbackAPI`:\n\n```python\nfrom semantic_lexicon import IntentClassifier, IntentExample\nfrom semantic_lexicon.api import FeedbackAPI, FeedbackService\nfrom semantic_lexicon.utils import read_jsonl\n\nexamples = [\n IntentExample(text=str(rec[\"text\"]), intent=str(rec[\"intent\"]), feedback=0.92)\n for rec in read_jsonl(\"src/semantic_lexicon/data/intent.jsonl\")\n]\nclassifier = IntentClassifier()\nclassifier.fit(examples)\nservice = FeedbackService(classifier)\napi = FeedbackAPI(service, host=\"127.0.0.1\", port=8765)\napi.start()\n```\n\nSubmit streaming feedback with a simple POST request:\n\n```bash\ncurl -X POST http://127.0.0.1:8765/feedback \\\n -H \"Content-Type: application/json\" \\\n -d '{\"prompt\": \"Compare supervised and unsupervised learning\", \\\n \"selected_intent\": \"comparison\", \\\n \"optimal_intent\": \"comparison\", \\\n \"feedback\": 0.96}'\n```\n\nThe server replies with the updated composite-reward weights and the component\nvector that was logged. Each event is processed under a lock so parallel clients\ncan stream feedback without clobbering the learned weights, and the new reward\nweights remain simplex-projected for EXP3 compatibility.\n\nKey parameters for `semantic-lexicon generate`:\n\n- `--workspace PATH` \u2013 directory that contains the trained embeddings and weights (defaults to `artifacts`).\n- `--persona NAME` \u2013 persona to blend into the response (defaults to the configuration's `default_persona`).\n- `--config PATH` \u2013 optional configuration file to override model hyperparameters during loading.\n\n## Adversarial Style Selection\n\nSemantic Lexicon now bundles EXP3 helpers for experimenting with\nadversarial persona *and* intent selection. The following snippet alternates\nbetween two personas while learning from scalar feedback in ``[0, 1]``:\n\n```python\nfrom semantic_lexicon import AnytimeEXP3, NeuralSemanticModel, SemanticModelConfig\nfrom semantic_lexicon.training import Trainer, TrainerConfig\n\nconfig = SemanticModelConfig()\nmodel = NeuralSemanticModel(config)\ntrainer = Trainer(model, TrainerConfig())\ntrainer.train()\n\nbandit = AnytimeEXP3(num_arms=2)\npersonas = [\"tutor\", \"researcher\"]\n\nfor prompt in [\n \"Outline matrix factorisation for recommendations\",\n \"Give journaling prompts about creativity\",\n \"Explain reinforcement learning trade-offs\",\n]:\n arm = bandit.select_arm()\n persona = personas[arm]\n response = model.generate(prompt, persona=persona)\n score = min(1.0, len(response.response.split()) / 40.0)\n bandit.update(score)\n```\n\n### Intent Selection with EXP3\n\nWe can model intent routing as an adversarial bandit problem. Let ``K`` be\nthe number of intents (e.g. ``{\"how_to\", \"definition\", \"comparison\", \"exploration\"}``).\nAt round ``t`` the system receives a prompt ``P_t`` and chooses an intent ``I_t``\nusing EXP3. After delivering the answer, a reward ``r_t`` in ``[0, 1]`` arrives\nfrom explicit ratings or engagement metrics. The arm-selection probabilities are\n\n$$\np_i(t) = (1 - \\gamma) \\frac{w_i(t)}{\\sum_{j=1}^{K} w_j(t)} + \\frac{\\gamma}{K},\n$$\n\nand the weight for the played intent updates via\n\n$$\nw_{I_t}(t+1) = w_{I_t}(t) \\exp\\left(\\frac{\\gamma r_t}{K p_{I_t}(t)}\\right).\n$$\n\nWhen the horizon ``T`` is unknown, the bundled ``AnytimeEXP3`` class applies the\ndoubling trick to refresh its parameters so the regret remains ``O(\\sqrt{T})``.\n\nThe quickstart script demonstrates the pattern by mapping arms to intent labels\nand simulating rewards from the classifier's posterior probability:\n\n```python\nfrom semantic_lexicon import AnytimeEXP3, NeuralSemanticModel, SemanticModelConfig\nfrom semantic_lexicon.training import Trainer, TrainerConfig\n\nconfig = SemanticModelConfig()\nmodel = NeuralSemanticModel(config)\ntrainer = Trainer(model, TrainerConfig())\ntrainer.train()\n\nintents = [label for _, label in sorted(model.intent_classifier.index_to_label.items())]\nbandit = AnytimeEXP3(num_arms=len(intents))\nprompt = \"How should I start researching renewable energy?\"\narm = bandit.select_arm()\nintent = intents[arm]\n reward = model.intent_classifier.predict_proba(prompt)[intent]\n bandit.update(reward)\n ```\n\n## Intent-Bandit Analysis Toolkit\n\nThe `semantic_lexicon.analysis` module supplies the maths underpinning the\nimproved EXP3 workflow:\n\n- `RewardComponents` & `composite_reward` combine correctness, calibration,\n semantic, and feedback signals into the bounded reward required by EXP3.\n- `estimate_optimal_weights` fits component weights via simplex-constrained least\n squares on historical interactions.\n- `DirichletCalibrator` provides Bayesian confidence calibration with a\n Dirichlet prior, yielding posterior predictive probabilities that minimise\n expected calibration error.\n- `simulate_intent_bandit` and `exp3_expected_regret` numerically check the\n \\(2.63\\sqrt{K T \\log K}\\) regret guarantee for the composite reward.\n- `compute_confusion_correction` and `confusion_correction_residual` extract the\n SVD-based pseudoinverse that reduces systematic routing errors.\n- `RobbinsMonroProcess` and `convergence_rate_bound` expose the stochastic\n approximation perspective with an \\(O(1/\\sqrt{n})\\) convergence rate bound.\n\nSee [docs/analysis.md](docs/analysis.md) for full derivations and proofs.\n\n### Intent Classification Objective\n\nEthical deployment requires robust intent understanding. Semantic Lexicon's\n``IntentClassifier`` treats intent prediction as a multinomial logistic regression\nproblem over prompts ``(P_i, I_i)``. Given parameters ``\\theta``, the model\nminimises the cross-entropy loss\n\n$$\n\\mathcal{L}(\\theta) = -\\frac{1}{N} \\sum_{i=1}^{N} \\log p(I_i \\mid P_i; \\theta),\n$$\n\nwhich matches the negative log-likelihood optimised during training. Improving\nintent accuracy directly translates into higher-quality feedback for the bandit\nloop.\n\n## Configuration\n\nSemantic Lexicon reads configuration files in YAML or JSON using the `SemanticModelConfig` dataclass. Example `config.yaml`:\n\n```yaml\nembeddings:\n dimension: 50\n max_words: 5000\nintent:\n learning_rate: 0.2\n epochs: 5\nknowledge:\n max_relations: 4\npersona:\n default_persona: tutor\ngenerator:\n temperature: 0.7\n```\n\nLoad the configuration via CLI (`semantic-lexicon train --config config.yaml`) or programmatically:\n\n```python\nfrom semantic_lexicon import NeuralSemanticModel, load_config\n\nconfig = load_config(\"config.yaml\")\nmodel = NeuralSemanticModel(config)\n```\n\n## Training API\n\n```python\nfrom semantic_lexicon import NeuralSemanticModel, SemanticModelConfig\nfrom semantic_lexicon.training import Trainer, TrainerConfig\n\nconfig = SemanticModelConfig()\nmodel = NeuralSemanticModel(config)\ntrainer = Trainer(model, TrainerConfig())\ntrainer.train()\nresponse = model.generate(\"How to learn python?\", persona=\"tutor\")\nprint(response.response)\n```\n\n## Diagnostics Programmatically\n\n```python\nfrom semantic_lexicon.model import NeuralSemanticModel\nfrom semantic_lexicon.training import Trainer, TrainerConfig\n\nmodel = NeuralSemanticModel()\ntrainer = Trainer(model, TrainerConfig())\ntrainer.train()\nreport = trainer.run_diagnostics()\nprint(report.to_dict())\n```\n\n## Development Workflow\n\n| Task | Command |\n| --------------- | --------------------------------- |\n| Format & lint | `ruff check .` \u00b7 `black .` |\n| Type check | `mypy src` |\n| Run tests | `pytest` |\n| Preview docs | `mkdocs serve` |\n\nA `Makefile` (or CI workflow) can orchestrate the tasks:\n\n```bash\nmake lint\nmake test\nmake docs\n```\n\n## Streams & Clipboard\n\nGeneration now distinguishes abstract sources via the prompt functor \\(\ud835\udc05\\). Use cases:\n- **Literal prompts** \u2013 pass a string and the CLI behaves exactly as before.\n- **Streaming prompts** \u2013 pass `\"-\"` to fold STDIN chunks until EOF, perfect for shell pipelines.\n- **Clipboard prompts** \u2013 call `semantic-lexicon clipboard` to pull the current system clipboard.\n\nExample invocations:\n\n```bash\necho \"What is a transformer?\" | semantic-lexicon generate - --workspace artifacts\nsemantic-lexicon clipboard --workspace artifacts --persona exploration\n```\n\nBoth paths reuse the existing workspace/persona/config pipeline and reject empty inputs with a friendly error.\n\nSample outputs:\n\n```text\n$ echo \"What is a transformer?\" | semantic-lexicon generate - --workspace artifacts\nPersona: generic\nResponse:\n1. Transformers rely on self-attention so tokens draw context from the entire sentence in one step.\n2. Multi-head attention lets the model track different relationships (syntax, long-range cues) simultaneously.\n3. Decoder layers reuse the same mechanism to generate fluent text token by token.\n\n$ semantic-lexicon clipboard --workspace artifacts --persona exploration\nClipboard prompt: \"Give me three research angles on causal discovery.\"\nPersona: exploration\nResponse:\n1. Explore score-based causal discovery that leverages diffusion models to recover graph structure from noise.\n2. Compare invariant risk minimisation versus meta-learning for handling interventions and domain shift.\n3. Prototype active experimentation loops that query the system for the most informative interventions next.\n```\n\n---\n\n## Contributing\n\n1. Fork the repository and create a feature branch.\n2. Install development dependencies: `pip install .[dev]`.\n3. Run `make test` to ensure linting, typing, and tests pass.\n4. Submit a pull request with detailed notes on new features or fixes.\n\n## Acknowledgments\n\nThis work was shaped by the survey \"Interpretation of Time-Series Deep Models: A Survey\" [(arXiv:2305.14582)](https://arxiv.org/abs/2305.14582) shared by Dr. Zhao after reading our preprint on Calibrated \"Counterfactual Conformal Fairness\" (C3F) [(arXiv:2509.25295)](https://arxiv.org/abs/2509.25295). His survey offered both the conceptual framing and motivation for exploring this research path. We also thank Hamdi Alakkad and Bugra Kilictas for their pivotal contributions to our related preprints, which laid the groundwork for the developments presented here. We further acknowledge DeepSeek, whose advanced mathematical reasoning and logical inference capabilities substantially enhanced the precision and efficiency of the formal logic analysis, and the collaboration between OpenAI and GitHub on Codex, whose code generation strengths, in concert with DeepSeek\u2019s systems, significantly accelerated and sharpened the overall development and analysis process.\n\n## Author's Note\nHello people, or a system running perfectly, inbetween or broken -- At least working. -- While I am building groups, it is nice to see you behind them. This project represents my core self. We all came from a fixed point and would end up there as well. I am working on making myself \u201cus,\u201d me \u201cour.\u201d The physical world is for receiving and giving feelings, while the symbolic world is the projection of those feelings. Today is October 13, 2025, and I am located in Meckenheim, Germany. My plane landed yesterday from Istanbul\u2014a nice trip, though (p.s. @farukalpayy). So, did you all feel like the energy was broken? It was the point where you get deep enough to realize where it was going. We reached the point where f(x) = x holds, but f(x) = y itself is also a point. And at this point, my request could be clarified. If this project saves you time or money, please consider sponsoring. Most importantly, it helps me keep improving and offering it free for the community. [Visit my Donation Page](https://buymeacoffee.com/farukalpay)\n\n## Contact & Legal\n\n- Semantic Lexicon is a Lightcap\u00ae research project distributed as open source under the Apache License 2.0; see [LICENSE](LICENSE) for details on rights and obligations.\n- Lightcap\u00ae is a registered trademark (EUIPO Reg. No. 019172085).\n- For enquiries, contact [alpay@lightcap.ai](mailto:alpay@lightcap.ai).\n",
"bugtrack_url": null,
"license": "Apache License\n Version 2.0, January 2004\n http://www.apache.org/licenses/\n \n TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION\n \n 1. Definitions.\n \n \"License\" shall mean the terms and conditions for use, reproduction,\n and distribution as defined by Sections 1 through 9 of this document.\n \n \"Licensor\" shall mean the copyright owner or entity authorized by\n the copyright owner that is granting the License.\n \n \"Legal Entity\" shall mean the union of the acting entity and all\n other entities that control, are controlled by, or are under common\n control with that entity. For the purposes of this definition,\n \"control\" means (i) the power, direct or indirect, to cause the\n direction or management of such entity, whether by contract or\n otherwise, or (ii) ownership of fifty percent (50%) or more of the\n outstanding shares, or (iii) beneficial ownership of such entity.\n \n \"You\" (or \"Your\") shall mean an individual or Legal Entity\n exercising permissions granted by this License.\n \n \"Source\" form shall mean the preferred form for making modifications,\n including but not limited to software source code, documentation\n source, and configuration files.\n \n \"Object\" form shall mean any form resulting from mechanical\n transformation or translation of a Source form, including but\n not limited to compiled object code, generated documentation,\n and conversions to other media types.\n \n \"Work\" shall mean the work of authorship, whether in Source or\n Object form, made available under the License, as indicated by a\n copyright notice that is included in or attached to the work\n (an example is provided in the Appendix below).\n \n \"Derivative Works\" shall mean any work, whether in Source or Object\n form, that is based on (or derived from) the Work and for which the\n editorial revisions, annotations, elaborations, or other modifications\n represent, as a whole, an original work of authorship. For the purposes\n of this License, Derivative Works shall not include works that remain\n separable from, or merely link (or bind by name) to the interfaces of,\n the Work and Derivative Works thereof.\n \n \"Contribution\" shall mean any work of authorship, including\n the original version of the Work and any modifications or additions\n to that Work or Derivative Works thereof, that is intentionally\n submitted to Licensor for inclusion in the Work by the copyright owner\n or by an individual or Legal Entity authorized to submit on behalf of\n the copyright owner. For the purposes of this definition, \"submitted\"\n means any form of electronic, verbal, or written communication sent\n to the Licensor or its representatives, including but not limited to\n communication on electronic mailing lists, source code control systems,\n and issue tracking systems that are managed by, or on behalf of, the\n Licensor for the purpose of discussing and improving the Work, but\n excluding communication that is conspicuously marked or otherwise\n designated in writing by the copyright owner as \"Not a Contribution.\"\n \n \"Contributor\" shall mean Licensor and any individual or Legal Entity\n on behalf of whom a Contribution has been received by Licensor and\n subsequently incorporated within the Work.\n \n 2. Grant of Copyright License. Subject to the terms and conditions of\n this License, each Contributor hereby grants to You a perpetual,\n worldwide, non-exclusive, no-charge, royalty-free, irrevocable\n copyright license to reproduce, prepare Derivative Works of,\n publicly display, publicly perform, sublicense, and distribute the\n Work and such Derivative Works in Source or Object form.\n \n 3. Grant of Patent License. Subject to the terms and conditions of\n this License, each Contributor hereby grants to You a perpetual,\n worldwide, non-exclusive, no-charge, royalty-free, irrevocable\n (except as stated in this section) patent license to make, have made,\n use, offer to sell, sell, import, and otherwise transfer the Work,\n where such license applies only to those patent claims licensable\n by such Contributor that are necessarily infringed by their\n Contribution(s) alone or by combination of their Contribution(s)\n with the Work to which such Contribution(s) was submitted. If You\n institute patent litigation against any entity (including a\n cross-claim or counterclaim in a lawsuit) alleging that the Work\n or a Contribution incorporated within the Work constitutes direct\n or contributory patent infringement, then any patent licenses\n granted to You under this License for that Work shall terminate\n as of the date such litigation is filed.\n \n 4. Redistribution. You may reproduce and distribute copies of the\n Work or Derivative Works thereof in any medium, with or without\n modifications, and in Source or Object form, provided that You\n meet the following conditions:\n \n (a) You must give any other recipients of the Work or\n Derivative Works a copy of this License; and\n \n (b) You must cause any modified files to carry prominent notices\n stating that You changed the files; and\n \n (c) You must retain, in the Source form of any Derivative Works\n that You distribute, all copyright, patent, trademark, and\n attribution notices from the Source form of the Work,\n excluding those notices that do not pertain to any part of\n the Derivative Works; and\n \n (d) If the Work includes a \"NOTICE\" text file as part of its\n distribution, then any Derivative Works that You distribute must\n include a readable copy of the attribution notices contained\n within such NOTICE file, excluding those notices that do not\n pertain to any part of the Derivative Works, in at least one\n of the following places: within a NOTICE text file distributed\n as part of the Derivative Works; within the Source form or\n documentation, if provided along with the Derivative Works; or,\n within a display generated by the Derivative Works, if and\n wherever such third-party notices normally appear. The contents\n of the NOTICE file are for informational purposes only and\n do not modify the License. You may add Your own attribution\n notices within Derivative Works that You distribute, alongside\n or as an addendum to the NOTICE text from the Work, provided\n that such additional attribution notices cannot be construed\n as modifying the License.\n \n You may add Your own copyright statement to Your modifications and\n may provide additional or different license terms and conditions\n for use, reproduction, or distribution of Your modifications, or\n for any such Derivative Works as a whole, provided Your use,\n reproduction, and distribution of the Work otherwise complies with\n the conditions stated in this License.\n \n 5. Submission of Contributions. Unless You explicitly state otherwise,\n any Contribution intentionally submitted for inclusion in the Work\n by You to the Licensor shall be under the terms and conditions of\n this License, without any additional terms or conditions.\n Notwithstanding the above, nothing herein shall supersede or modify\n the terms of any separate license agreement you may have executed\n with Licensor regarding such Contributions.\n \n 6. Trademarks. This License does not grant permission to use the trade\n names, trademarks, service marks, or product names of the Licensor,\n except as required for reasonable and customary use in describing the\n origin of the Work and reproducing the content of the NOTICE file.\n \n 7. Disclaimer of Warranty. Unless required by applicable law or\n agreed to in writing, Licensor provides the Work (and each\n Contributor provides its Contributions) on an \"AS IS\" BASIS,\n WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or\n implied, including, without limitation, any warranties or conditions\n of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A\n PARTICULAR PURPOSE. You are solely responsible for determining the\n appropriateness of using or redistributing the Work and assume any\n risks associated with Your exercise of permissions under this License.\n \n 8. Limitation of Liability. In no event and under no legal theory,\n whether in tort (including negligence), contract, or otherwise,\n unless required by applicable law (such as deliberate and grossly\n negligent acts) or agreed to in writing, shall any Contributor be\n liable to You for damages, including any direct, indirect, special,\n incidental, or consequential damages of any character arising as a\n result of this License or out of the use or inability to use the\n Work (including but not limited to damages for loss of goodwill,\n work stoppage, computer failure or malfunction, or any and all\n other commercial damages or losses), even if such Contributor\n has been advised of the possibility of such damages.\n \n 9. Accepting Warranty or Additional Liability. While redistributing\n the Work or Derivative Works thereof, You may choose to offer,\n and charge a fee for, acceptance of support, warranty, indemnity,\n or other liability obligations and/or rights consistent with this\n License. However, in accepting such obligations, You may act only\n on Your own behalf and on Your sole responsibility, not on behalf\n of any other Contributor, and only if You agree to indemnify,\n defend, and hold each Contributor harmless for any liability\n incurred by, or claims asserted against, such Contributor by reason\n of your accepting any such warranty or additional liability.\n \n END OF TERMS AND CONDITIONS\n \n APPENDIX: How to apply the Apache License to your work.\n \n To apply the Apache License to your work, attach the following\n boilerplate notice, with the fields enclosed by brackets \"[]\"\n replaced with your own identifying information. (Don't include\n the brackets!) The text should be enclosed in the appropriate\n comment syntax for the file format. We also recommend that a\n file or class name and description of purpose be included on the\n same \"printed page\" as the copyright notice for easier\n identification within third-party archives.\n \n Copyright [yyyy] [name of copyright owner]\n \n Licensed under the Apache License, Version 2.0 (the \"License\");\n you may not use this file except in compliance with the License.\n You may obtain a copy of the License at\n \n http://www.apache.org/licenses/LICENSE-2.0\n \n Unless required by applicable law or agreed to in writing, software\n distributed under the License is distributed on an \"AS IS\" BASIS,\n WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n See the License for the specific language governing permissions and\n limitations under the License.\n ",
"summary": "Neural semantic modeling toolkit with persona-aware generation and diagnostics.",
"version": "0.1.4",
"project_urls": {
"Documentation": "https://github.com/farukalpay/Semantic-Lexicon#readme",
"Homepage": "https://github.com/farukalpay/Semantic-Lexicon",
"Issues": "https://github.com/farukalpay/Semantic-Lexicon/issues"
},
"split_keywords": [
"nlp",
" semantics",
" persona",
" knowledge-graph",
" cli"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "586f8a5a7c88c5e23c04d5e63fadfbe3a7ce8b1c5b28a57ebf416341e46a48a3",
"md5": "adb7f814f11dc57f3ae14e0d2b16abfd",
"sha256": "0d45455002a6eef895ac8ffee7919632e1ed5adf8317a0b474cc59b0a003f1f9"
},
"downloads": -1,
"filename": "semantic_lexicon-0.1.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "adb7f814f11dc57f3ae14e0d2b16abfd",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 171762,
"upload_time": "2025-10-19T22:28:42",
"upload_time_iso_8601": "2025-10-19T22:28:42.244170Z",
"url": "https://files.pythonhosted.org/packages/58/6f/8a5a7c88c5e23c04d5e63fadfbe3a7ce8b1c5b28a57ebf416341e46a48a3/semantic_lexicon-0.1.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "6711c53b3621ae21353320f462c670fdf809e6475c6c359aca96b99d0d83f645",
"md5": "47846f1c992412cb0cc9ace6363d9389",
"sha256": "39b6cc1eed5c570ed1a77cd20050bb974751950b9d868632434f387ddc534e5d"
},
"downloads": -1,
"filename": "semantic_lexicon-0.1.4.tar.gz",
"has_sig": false,
"md5_digest": "47846f1c992412cb0cc9ace6363d9389",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 204537,
"upload_time": "2025-10-19T22:28:43",
"upload_time_iso_8601": "2025-10-19T22:28:43.809524Z",
"url": "https://files.pythonhosted.org/packages/67/11/c53b3621ae21353320f462c670fdf809e6475c6c359aca96b99d0d83f645/semantic_lexicon-0.1.4.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-10-19 22:28:43",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "farukalpay",
"github_project": "Semantic-Lexicon#readme",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "semantic-lexicon"
}
