We propose OPERA (Offline Policy-guided Expert Routing and Adaptation), a multi-agent ensemble framework that addresses this deployment bottleneck by treating expert weight assignment as an offline policy learning problem: a routing policy is learned from a small validation set without gradient updates to any expert agent, then deployed with test-time adaptation to handle distribution shift. OPERA coordinates heterogeneous specialist agents through complementary mechanisms. The expert profiling module learns selection policies offline, enabling informed allocation of expertise. Each agent undergoes confidence calibration through temperature adjustment, ensuring more reliable probabilistic outputs.</p>\n","updatedAt":"2026-07-29T17:29:58.109Z","author":{"_id":"64176d65b03817ada642b6e2","avatarUrl":"/avatars/27fe82eaf9db2894c0bf86acb59543fe.svg","fullname":"Zihan Li","name":"zl111","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":7,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8637775778770447},"editors":["zl111"],"editorAvatarUrls":["/avatars/27fe82eaf9db2894c0bf86acb59543fe.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.25108","authors":[{"_id":"6a6a38474463a8a84bdc3e5d","name":"Zihan Li","hidden":false},{"_id":"6a6a38474463a8a84bdc3e5e","name":"Feiyang Liu","hidden":false},{"_id":"6a6a38474463a8a84bdc3e5f","name":"Dandan Shan","hidden":false},{"_id":"6a6a38474463a8a84bdc3e60","name":"Ruibo Wang","hidden":false},{"_id":"6a6a38474463a8a84bdc3e61","name":"Qingqi Hong","hidden":false}],"publishedAt":"2026-07-27T00:00:00.000Z","submittedOnDailyAt":"2026-07-29T00:00:00.000Z","title":"OPERA: Offline Policy-guided Expert Routing and Adaptation for Universal Biomedical Image Analysis","submittedOnDailyBy":{"_id":"64176d65b03817ada642b6e2","avatarUrl":"/avatars/27fe82eaf9db2894c0bf86acb59543fe.svg","isPro":false,"fullname":"Zihan Li","user":"zl111","type":"user","name":"zl111"},"summary":"Biomedical image analysis spans diverse modalities and tasks, yet real-world deployment is hindered by severe distribution shifts across scanners, protocols, and patient populations. High-performing models consequently require repeated domain-specific fine-tuning, which is a costly cycle that becomes impractical when labels are scarce or privacy constraints limit data sharing. We propose OPERA (Offline Policy-guided Expert Routing and Adaptation), a multi-agent ensemble framework that addresses this deployment bottleneck by treating expert weight assignment as an offline policy learning problem: a routing policy is learned from a small validation set without gradient updates to any expert agent, then deployed with test-time adaptation to handle distribution shift. OPERA coordinates heterogeneous specialist agents through complementary mechanisms. The expert profiling module learns selection policies offline, enabling informed allocation of expertise. Each agent undergoes confidence calibration through temperature adjustment, ensuring more reliable probabilistic outputs. OPERA also incorporates distribution aware adaptation, where class weights are dynamically adjusted at the batch level using statistics derived from unlabeled test data. Instance level routing assigns each sample to the most suitable expert by leveraging inter model agreement and predictive entropy. We evaluate OPERA on 9 datasets covering fundus photography, chest X-ray, CT, MRI, and multimodal diagnostic benchmarks, comparing against 30+ baselines across classification, segmentation, and multimodal settings. OPERA consistently improves performance and calibration quality, demonstrating that offline policy-guided expert agents coordination is a practical path to deployable biomedical AI without retraining. Code is on https://github.com/HUANGLIZI/OPERA{GitHub}.","upvotes":0,"discussionId":"6a6a38474463a8a84bdc3e62","projectPage":"https://github.com/HUANGLIZI/OPERA","githubRepo":"https://github.com/HUANGLIZI/OPERA","githubRepoAddedBy":"user","githubStars":1,"organization":{"_id":"6315a1bb86b3db2ac420100e","name":"UW","fullname":"University of Washington","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/gr5B_WVvbMr4kTox5UkwZ.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"6315a1bb86b3db2ac420100e","name":"UW","fullname":"University of Washington","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/gr5B_WVvbMr4kTox5UkwZ.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.25108.md","query":{}}">
OPERA: Offline Policy-guided Expert Routing and Adaptation for Universal Biomedical Image Analysis
Abstract
Biomedical image analysis spans diverse modalities and tasks, yet real-world deployment is hindered by severe distribution shifts across scanners, protocols, and patient populations. High-performing models consequently require repeated domain-specific fine-tuning, which is a costly cycle that becomes impractical when labels are scarce or privacy constraints limit data sharing. We propose OPERA (Offline Policy-guided Expert Routing and Adaptation), a multi-agent ensemble framework that addresses this deployment bottleneck by treating expert weight assignment as an offline policy learning problem: a routing policy is learned from a small validation set without gradient updates to any expert agent, then deployed with test-time adaptation to handle distribution shift. OPERA coordinates heterogeneous specialist agents through complementary mechanisms. The expert profiling module learns selection policies offline, enabling informed allocation of expertise. Each agent undergoes confidence calibration through temperature adjustment, ensuring more reliable probabilistic outputs. OPERA also incorporates distribution aware adaptation, where class weights are dynamically adjusted at the batch level using statistics derived from unlabeled test data. Instance level routing assigns each sample to the most suitable expert by leveraging inter model agreement and predictive entropy. We evaluate OPERA on 9 datasets covering fundus photography, chest X-ray, CT, MRI, and multimodal diagnostic benchmarks, comparing against 30+ baselines across classification, segmentation, and multimodal settings. OPERA consistently improves performance and calibration quality, demonstrating that offline policy-guided expert agents coordination is a practical path to deployable biomedical AI without retraining. Code is on https://github.com/HUANGLIZI/OPERA{GitHub}.
Community
We propose OPERA (Offline Policy-guided Expert Routing and Adaptation), a multi-agent ensemble framework that addresses this deployment bottleneck by treating expert weight assignment as an offline policy learning problem: a routing policy is learned from a small validation set without gradient updates to any expert agent, then deployed with test-time adaptation to handle distribution shift. OPERA coordinates heterogeneous specialist agents through complementary mechanisms. The expert profiling module learns selection policies offline, enabling informed allocation of expertise. Each agent undergoes confidence calibration through temperature adjustment, ensuring more reliable probabilistic outputs.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.25108 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.25108 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.25108 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.