HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Thu, 03 Jul 2025 21:56:41 GMT access-control-allow-origin: * strict-transport-security: max-age=31556952 etag: W/"6866fc99-fa0" expires: Mon, 29 Dec 2025 16:06:15 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: 26F9:2685F2:8F855D:A123CF:6952A49F accept-ranges: bytes age: 0 date: Mon, 29 Dec 2025 15:56:16 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210056-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767023776.851354,VS0,VE218 vary: Accept-Encoding x-fastly-request-id: a1720fc393662ccc187294314138247c8838e38a content-length: 1408

BehaviorSFT: Behavioral Token Conditioning for Clinical Agents Across the Proactivity Spectrum

📁 SFT Code 📁 Model Checkpoints

Six representative tasks from BehaviorBench, showcasing the spectrum of agent behaviors in clinical settings. The figure illustrates (a-c, f) proactive tasks where the LLM agent identifies issues or offers insights without direct prompting, and (b, d, e) reactive tasks responding to explicit clinician queries