Downloads · 30 days
7
14% of all-time downloads
n0rths7ar/bioevidence-qwen3-4b-extraction-lora-v1
bioevidence-qwen3-4b-extraction-lora-v1 is a text generation model from n0rths7ar. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This is a QLoRA adapter for query-focused structured extraction from one PubMed title and abstract. It was trained as part of BioEvidence Copilot and targets the repository's versioned ModelEvidenceExtraction JSON Sch…
Downloads · 30 days
7
14% of all-time downloads
All-time downloads
50
Public
Repo size
77.6 MB
Likes
1
Public
Click a slice to open those files.
.safetensors66.1 MB · 85%
From the Hugging Face model README
This is a QLoRA adapter for query-focused structured extraction from one PubMed
title and abstract. It was trained as part of
BioEvidence Copilot and
targets the repository's versioned ModelEvidenceExtraction JSON Schema.
The adapter is an engineering experiment in strict structured output and grounded evidence spans. It is not a biomedical decision model.
Input consists of a biomedical query, a PubMed title, an abstract, and the v1 JSON Schema. Output is exactly one JSON object containing:
Use the adapter for reproducible portfolio experiments, offline extraction evaluation, and optional enrichment inside BioEvidence Copilot. Validate every response against the schema and verify evidence spans against the source text.
Do not use it for diagnosis, treatment selection, clinical decision-making, or unsupported claims beyond the supplied title and abstract.
unsloth/Qwen3-4B-Instruct-2507-unsloth-bnb-4bit7744afa8566e264af1a92a806d8d9aae00cc7c78The v1 experiment contains 60 query-document annotations built from PubMed titles and abstracts:
| Label | Rows |
|---|---|
| direct | 8 |
| indirect | 29 |
| none | 23 |
Labels are model-assisted drafts. The 20-row pilot received a second-model review; 40 expansion rows were produced in Claude Desktop and passed JSON, Pydantic schema, duplicate-pair, and exact-span validation. Splits are assigned by PMID to prevent the same document from crossing train, dev, and test.
| Split | Rows | Unique PMIDs |
|---|---|---|
| train | 46 | 43 |
| dev | 7 | 5 |
| test | 7 | 5 |
Dataset provenance and split IDs are tracked in the project repository. The adapter package does not redistribute the training dataset.
{All systems were evaluated on the same seven PMID-held-out draft annotations with deterministic decoding.
| Metric | Rules | Prompted base | QLoRA adapter |
|---|---|---|---|
| Strict JSON parse rate | 1.000 | 1.000 | 1.000 |
| Schema validity | 1.000 | 1.000 | 1.000 |
| Grounding rate | n/a | 0.714 | 1.000 |
| Evidence-status accuracy | 0.429 | 0.714 | 0.429 |
| Study-design accuracy | 0.857 | 1.000 | 0.857 |
| Semantic-field token F1 | 0.571 | 0.540 | 0.668 |
| Outcome-name token F1 | 0.429 | 0.414 | 0.631 |
| Outcome-direction accuracy | 0.429 | 0.536 | 0.714 |
| Evidence-span token F1 | 0.513 | 0.461 | 0.511 |
| Evidence-span support rate | 1.000 | 0.857 | 1.000 |
| Mean generation time | n/a | 20.08 s | 7.34 s |
These results demonstrate strict JSON specialization and a working local training/inference path. Seven draft test rows are too few for a general claim about biomedical extraction quality, and evidence-status classification did not improve over the prompted base model.
The repository contains the exact environment lock, schema, annotations, split manifest, training script, evaluator, and aggregate report. After placing the adapter in a local directory:
conda activate bioevidence-training
python scripts/run_extraction_eval.py `
--backend local `
--adapter-path path/to/adapter `
--dataset artifacts/training/evidence_extraction/training_v1_sft/test.annotations.jsonl
For optional product inference, set EXTRACTION_BACKEND=local and
EXTRACTION_ADAPTER_PATH=path/to/adapter. BioEvidence Copilot validates schema
and verbatim spans and falls back to its deterministic structured extractor when
the optional local runtime is unavailable or invalid.