Automated README summary
Automatically summarized from the README on 25 Sept 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
Jev-MedQA selects answers to medical text and image questions from candidate options in one forward pass, using Qwen3.5-4B without additional training. It is for medical QA benchmarks and clinical decision support.
How it uses Jev
Jev-MedQA reads the model's logits for each candidate answer letter (A-D) in a single forward pass over the question and any images, then returns the highest-scoring option. Softmax converts logits to relative scores. The choice and scores are used directly as the answer.
Primitives:choice
Technique worth stealing
Using logits of answer letters as a direct choice mechanism, avoiding token generation.
Try it
conda create -n jev-medqa python=3.10 -y && conda activate jev-medqa && pip install -r requirements.txt && python scripts/evaluate.py --gpu-ids 0,1 --model /path/to/Qwen3.5-4B --inputs questions.jsonl --answers answers.jsonl --output runs/full
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectFound in material0.86
- Shows a System One patternnot found in the excerpt0.33
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersFound in material0.98
- Is a runnable toolnot found in the excerpt0.30
- Top-10 learning shortlist signalunclear0.53
- Is a replica of the modelFound in material0.93
- Solves a common problem (0–2)score on a 0–2 scale1.02 / 2
- About JevFound in material0.92
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 25 Sept 2026.