Update README.md

README.md

@@ -1,23 +1,21 @@
----
-license: mit
-base_model:
-- Qwen/Qwen3-VL-8B-Instruct
----
+# case-8b-rl (step 1912, 1 epoch)
 
-# case-8b-rl (step 1912，1 epoch)
-
-**Final RL checkpoint** of CASE - a multi-turn tool-use RL agent for medical imaging.
+**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*.
 
-- **Training**: GRPO + OPSD (bucket-IV causal-KL) + rubric-judge reward; multi-turn
-  tool rollout over three clinician roles (Anatomist / Radiologist / Consultant)
-- **Step**: 1912 (final; run `grpo_qwen3vl_tool_rl`)
-- **Provenance**: `case/rl_qwen3vl_tool/hf_step1912` (merged from FSDP shards)
+- **Training**: agentic RL - GRPO over multi-turn tool rollouts against a mock
+  tool-replay environment, + OPSD (bucket-IV causal-KL) + rubric-judge reward
+- **Step**: 1912 (final)
 - **Base model**: Qwen3-VL-8B-Instruct
-- **Task**: longitudinal medical-imaging QA; the policy answers by calling tools
-  (view image, compare visits, organ segmentation, report retrieval, skill memory)
 - **Format**: merged HF checkpoint (single `model.safetensors` + tokenizer/processor)
-- **Usage**: `vllm serve <this_dir> --trust-remote-code` (add `--enforce-eager` when
-  serving in an environment without a full CUDA toolkit)
+- **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.
+  data is redistributed here. Research use only.