Downloads · 30 days
17
16% of all-time downloads
vadimbelsky/qwen3.5-esi-triage-grpo-v49-merged
qwen3.5-esi-triage-grpo-v49-merged is a text generation model from vadimbelsky. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is the merged full-weights version of the v49 GRPO ESI triage adapter — an iteration on v46 that targets specific clinical-rule failure modes identified in v46's error analysis.
Downloads · 30 days
17
16% of all-time downloads
All-time downloads
109
Public
Parameters
9.4B
18.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors18.8 GB · 100%
From the Hugging Face model README
This is the merged full-weights version of the v49 GRPO ESI triage adapter — an iteration on v46 that targets specific clinical-rule failure modes identified in v46's error analysis.
Error triage on v46's 8 wrong cases on MIETIC-36 revealed a clear pattern:
v49 keeps everything that made v46 work (the 0.0 safety valve, asymmetric over-/under-triage penalties, 5× ESI-1 oversample) and adds rule-aware reward bonuses:
| Trigger pattern in case text | Bonus / penalty |
|---|---|
| `intubat | chest tube |
| `intubat | chest tube |
| `open fracture | penetrating |
| Pain ≥ 7 + gold=2 + pred>2 | -0.5 |
Other changes:
v49 was the cleanest GRPO run of the series:
clipped_ratio held at 4–19% throughout (vs v48's 90%+ for most of training)reward positive from step 10 (warm-start) with mid-run peak ~+0.49reward_std consistently 1.0+ — strong learning signal, no group collapsefrom transformers import AutoTokenizer, AutoModelForCausalLM
import torch
repo = "vadimbelsky/qwen3.5-esi-triage-grpo-v49-merged"
tokenizer = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype=torch.bfloat16, device_map="auto")
SYSTEM = (
"You are an expert emergency triage nurse. "
"Extract clinical fields, apply the ESI algorithm step by step, then state the ESI level. "
"Be concise — stay under 150 words total."
)
case = ("A 78-year-old female arrived intubated for airway protection after being found "
"unresponsive. Central line placed. BP 120/58, HR 150, RR 20, SpO2 97%. "
"Fever and presumed pneumonia. ")
prompt = tokenizer.apply_chat_template(
[{"role": "system", "content": SYSTEM},
{"role": "user", "content": case}],
tokenize=False, add_generation_prompt=True,
)
out = model.generate(
**tokenizer(prompt, return_tensors="pt").to(model.device),
max_new_tokens=1024, temperature=0.1, do_sample=True,
)
print(tokenizer.decode(out[0], skip_special_tokens=True))
Expected behavior on the example above: the model should now recognize that an already-intubated patient with a central line meets ESI-1 Step A criteria (lifesaving intervention performed), and answer ANSWER: ESI 1 — fixing one of v46's known failure cases.
Evaluation results vs v46 will be added once the MIETIC-36 dual-mode eval completes. v46 baseline: 77.8% exact / 94.4% adjacent.
Research model. Not approved for clinical use. Same caveats as v46. The rule bonuses target specific patterns we measured failing in v46 — they don't generalize to clinical rules outside that regex set. Cases that require rules outside {intubation, lifesaving intervention, open injury, severe pain} may still err.