---
library_name: transformers
tags:
- generated_from_trainer
- trl
- grpo
- hf_jobs
- clinical-recruitment
- openenv
- long-horizon
- lora
license: mit
base_model: Qwen/Qwen3-1.7B
---

# Clinical Recruitment GRPO Agent (Best Run — 80 Steps)

This is the best trained LoRA adapter for the [Adaptive Clinical Recruitment Environment](https://huggingface.co/spaces/pratimassaravanan/clinical-recruitment), produced by SFT warmup followed by 80-step GRPO on Qwen3-1.7B (NVIDIA L4, 24GB).

**Hackathon positioning: Theme #2 (Super Long-Horizon Planning).**

## Training Summary

| Metric | Value |
|--------|-------|
| Base model | Qwen/Qwen3-1.7B |
| Method | SFT warmup + 80-step GRPO |
| GPU | NVIDIA L4 (24GB) via HF Jobs |
| Duration | 142 min total (3.5 min SFT + 141.7 min GRPO) |
| Reward (start) | 0.269 |
| Reward (end) | 0.331 |
| Tool calls/step (start) | 3.5 |
| Tool calls/step (end) | 11 |
| Enrollment/rollout | 3-4 patients |
| Zero-std collapse rate | 10% (8/80 steps, intermittent) |
| LoRA rank | 16 |
| LoRA alpha | 16 |
| LoRA dropout | 0.05 |

## What the Model Learned

- Calls `screen_patient` as first action (previously collapsed to `adjust_strategy`)
- Follows the correct pipeline: screen -> recontact -> enrollment
- Enrolls 3-4 patients per episode on easy_bench
- Makes 11+ tool calls per rollout (up from 3.5)
- Recovers from intermittent collapse without sustained degradation

## Training Plots

![GRPO 80-Step Training](https://huggingface.co/pratimassaravanan/clinical-recruitment-artifacts/resolve/main/article_assets/grpo_80step_training.png)

![All Runs Comparison](https://huggingface.co/pratimassaravanan/clinical-recruitment-artifacts/resolve/main/article_assets/all_runs_comparison.png)

## Honest Limits

- Enrollment stays at 3-4/80 target patients per episode
- Zero-std collapse still occurs on ~10% of steps
- Reward plateaued at 0.33 after 80 steps

## Links

- **Live environment**: [pratimassaravanan-clinical-recruitment.hf.space](https://pratimassaravanan-clinical-recruitment.hf.space)
- **HF Space**: [pratimassaravanan/clinical-recruitment](https://huggingface.co/spaces/pratimassaravanan/clinical-recruitment)
- **Training script**: `train_sft_grpo_hfjob.py` in the Space repo
- **30-step first run**: [pratimassaravanan/grpo_output](https://huggingface.co/pratimassaravanan/grpo_output)
- **SFT+REINFORCE model**: [pratimassaravanan/clinical-qwen3-4b-sft-lora](https://huggingface.co/pratimassaravanan/clinical-qwen3-4b-sft-lora)
- **Artifacts**: [pratimassaravanan/clinical-recruitment-artifacts](https://huggingface.co/pratimassaravanan/clinical-recruitment-artifacts)

## Framework Versions

- TRL: 1.2.0
- Transformers: 5.6.2
- PyTorch: 2.11.0
- PEFT: 0.19.1