pyrrho-nano-g3.1

pyrrho-nano-g3.1 is a small multitask RAG governance co-processor for anti-hallucination and retrieval-quality pipelines. It reads a user question plus retrieved source passages, then returns a calibrated evidence-state decision and auxiliary signals that fitz-sage can use before answer generation.

It is not an answer generator and not an open-world fact checker. It sits between retrieval and generation, or beside a retrieval package as a fast evidence quality layer. Compared with pyrrho-nano-g3, this package adds multitask heads for pre-retrieval query-contract classification, semantic route/domain, taxonomy pattern, and six scalar governance signals.

Governance Labels

Label	Meaning
ABSTAIN	The retrieved sources do not contain enough evidence to answer the question.
DISPUTED	The retrieved sources conflict on the answer.
TRUSTWORTHY	The retrieved sources consistently support answering the question.

Multitask Heads

Head	Labels / values	Intended use
governance	ABSTAIN, DISPUTED, TRUSTWORTHY	Post-retrieval evidence sufficiency and conflict decision.
query_contract	evidence_sufficiency, structured_lookup, temporal_grounding, exhaustive_coverage, comparison_coverage, representative_overview	Pre-retrieval routing signal for what kind of evidence the query needs.
route	science_medicine, law_policy, history_geography, technology_computing, economics_finance, culture_society, general_commonsense	Semantic route/domain signal for retrieval policy and logging.
taxonomy	23 fitz-gov taxonomy patterns	Failure/support pattern signal for audit and diagnostics.
scalars	evidence_sufficiency, query_evidence_alignment, answer_coverage, conflict_density, retrieval_retry_value, false_trustworthy_risk	Continuous governance signals for retry, ranking, and monitoring.

Outputs

This is a custom multitask package, not a standard single-head AutoModelForSequenceClassification artifact. The recommended runtime is pyrrho.multitask_inference.PyrrhoMultiTaskPredictor from the pyrrho repository.

The predictor returns a structured object:

Field	Meaning
governance.final_label	Final calibrated label after the TRUSTWORTHY threshold rule.
governance.raw_label	Highest-probability governance label before threshold calibration.
governance.probabilities	Probability distribution over ABSTAIN, DISPUTED, TRUSTWORTHY.
governance.threshold	TRUSTWORTHY probability threshold used by the package.
query_contract.final_label	Query-only contract prediction.
route.final_label	Query-only semantic route/domain prediction.
taxonomy.final_label	Query+evidence taxonomy-pattern prediction.
scalars	Six bounded scalar governance signals.
timing_ms	Local inference timing for the call.

Example normalized output shape:

{
  "schema_version": "pyrrho_multitask_prediction_v1",
  "governance": {
    "raw_label": "TRUSTWORTHY",
    "final_label": "TRUSTWORTHY",
    "used_threshold_fallback": false,
    "threshold": 0.39,
    "confidence": 0.84,
    "probabilities": {
      "ABSTAIN": 0.08,
      "DISPUTED": 0.08,
      "TRUSTWORTHY": 0.84
    }
  },
  "query_contract": {
    "final_label": "structured_lookup"
  },
  "route": {
    "final_label": "economics_finance"
  },
  "taxonomy": {
    "final_label": "direct_answer"
  },
  "scalars": {
    "evidence_sufficiency": 0.91,
    "query_evidence_alignment": 0.88,
    "answer_coverage": 0.86,
    "conflict_density": 0.08,
    "retrieval_retry_value": 0.12,
    "false_trustworthy_risk": 0.09
  }
}

The model does not generate answers, citations, source spans, retrieval results, or natural-language explanations. It classifies and scores the (query, retrieved_contexts) evidence state.

Intended Use

Use this model when a RAG or retrieval package needs fast local signals about:

• whether retrieved evidence is enough to answer,
• whether retrieved evidence conflicts,
• what kind of evidence the query needs before retrieval,
• which semantic/domain route the query belongs to,
• which fitz-gov support/failure pattern is active,
• whether retrieval should retry, broaden, or escalate.

This model is not intended to write answers, verify facts outside the provided sources, replace a retriever, or replace human review in high-stakes settings.

Quick Start

Install the pyrrho package from the repository that contains this runtime, then load the package with the multitask predictor:

from huggingface_hub import snapshot_download

from pyrrho.multitask_inference import PyrrhoMultiTaskPredictor

MODEL_ID = "yafitzdev/pyrrho-nano-g3.1"
PACKAGE_DIR = snapshot_download(MODEL_ID)

query = "Which quarterly report is relevant?"
contexts = [
    "The Q2 report lists revenue, churn, and roadmap changes.",
]

predictor = PyrrhoMultiTaskPredictor.from_pretrained(PACKAGE_DIR, device="cpu")
result = predictor.predict(query, contexts)

print(result["governance"]["final_label"])
print(result["query_contract"]["final_label"])
print(result["route"]["final_label"])
print(result["taxonomy"]["final_label"])
print(result["scalars"])

For local package testing:

python scripts/package_multitask_encoder.py verify --package-dir models/pyrrho-nano-g3.1 --device cpu

Release Selection

• Seed: 7
• TRUSTWORTHY threshold: 0.39
• Selection reason: seed 7 had the strongest composite release score while retaining strong governance, query-contract, route, taxonomy, and scalar metrics.

Held-Out Test Metrics

Metric	Result
Governance accuracy	0.9805
False-TRUSTWORTHY rate	0.0095
Query-contract accuracy	0.9492
Query-contract macro F1	0.9423
Route accuracy	0.9296
Route macro F1	0.9282
Taxonomy accuracy	0.8943
Taxonomy macro F1	0.8960
Scalar MAE	0.0587

Three-seed headline from the local release summary:

Metric	Mean +/- std
Governance accuracy	97.84 +/- 0.15%
False-TRUSTWORTHY rate	0.85 +/- 0.07%
Query-contract macro F1	94.24 +/- 0.28%
Route accuracy	93.41 +/- 0.32%
Taxonomy accuracy	89.26 +/- 0.23%
Scalar MAE	0.0592 +/- 0.0005

Training Data

Trained on fitz-gov V8.1-style rows prepared from the V8.0.1 row set plus the mandatory routing.query_contract field. The release package records the local training config in training_config.yaml and detailed metrics in reports/summary.json.

Limitations

• This is a governance and routing co-processor, not a generator.
• The auxiliary heads are useful signals, not ground-truth explanations.
• Query-contract and route predictions are query-only and can be wrong when the user query is underspecified.
• Taxonomy and scalar outputs are trained on fitz-gov labels/signals and should be treated as decision-support metadata, not universal factual judgments.
• The license is CC BY-NC 4.0. Commercial use requires a separate license.