case-8b-rsft

Clinical Agents for Seeking Evidence - agentic harness learning for longitudinal medical imaging. The model autonomously plans and calls tools inside a multi-turn harness (model -> qwen3_xml tool call -> environment executes -> observation -> ... -> final answer); it never answers single-shot. Three role personas: Anatomist (organ segmentation / volumetry), Radiologist (cross-visit image comparison), Consultant (report-grounded synthesis).

Paper: From Given to Gathered Evidence: Agentic Learning for Longitudinal Medical Reasoning.

arxiv.org/abs/2609.39566

SFT-initialized agent (40% of the training split) - already tool-calling capable; this is the initialization that the RL models continue from.

• Training: mixed single-turn SFT over the three agent roles, assistant-only loss
• Step: final = 1515（1 epoch）
• Base model: Qwen3-VL-8B-Instruct
• Format: merged HF checkpoint (single model.safetensors + tokenizer/processor)
• Serving tip: run with a tool backend (the released mock-tool replay configs, or the full LangGraph agents) and start vLLM with --enable-auto-tool-choice --tool-call-parser qwen3_xml (add --enforce-eager when the environment has no full CUDA toolkit)
• Data: trained on UK Biobank-derived, access-controlled data; no UKB-derived data is redistributed here. Research use only.