IR3DE is a lightweight, modular, linear router that accurately matches each prompt to the most appropriate LLM expert.</p>\n","updatedAt":"2026-06-10T08:16:42.321Z","author":{"_id":"68e407dd07d44bad5e4e696b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/Ux50uRULEu5d_Ep4aYNNk.png","fullname":"Eros Fanì","name":"Erosinho","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":2,"identifiedLanguage":{"language":"en","probability":0.8969912528991699},"editors":["Erosinho"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/Ux50uRULEu5d_Ep4aYNNk.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.06098","authors":[{"_id":"6a291b66e7d78ea7587e56bd","name":"Eros Fanì","hidden":false},{"_id":"6a291b66e7d78ea7587e56be","name":"Oğuzhan Ersoy","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/68e407dd07d44bad5e4e696b/VKubKDLHEWenNOhfrvXzp.png"],"publishedAt":"2026-06-04T12:36:28.000Z","submittedOnDailyAt":"2026-06-10T00:00:00.000Z","title":"IR3DE: A Linear Router for Large Language Models","submittedOnDailyBy":{"_id":"68e407dd07d44bad5e4e696b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/Ux50uRULEu5d_Ep4aYNNk.png","isPro":false,"fullname":"Eros Fanì","user":"Erosinho","type":"user","name":"Erosinho"},"summary":"Foundational Large Language Models (LLMs) demonstrate proficiency on a wide range of general tasks, and achieve remarkable results on various specialized tasks via domain-expert LLMs. With the ever-growing list of available LLMs, inference routers are being proposed to select the most appropriate LLM for each prompt. However, existing routing methods either optimize cost across weak-to-strong generalist LLMs or require substantial training to support domain-expertise routing. In this paper, we propose IR3DE, a Ridge Regression-based Router for Domain Experts that provides cheap and fast routing decisions for each prompt. We evaluate IR3DE in two Causal Language Modeling (CLM) settings where the tasks are next-token prediction for all domains, and one reasoning setting where each domain has its own distinct reasoning task. Despite being a linear router, IR3DE achieves performance comparable to the other baselines in both CLM settings, and surpassing them in the reasoning setting, with a normalized performance of 98.4%. Moreover, IR3DE enables the addition or removal of new domain experts without requiring the router to be retrained from scratch, allowing a dynamic set of LLMs to be served with minimal disruption to the router itself. Our code is available at: github.com/gensyn-ai/IR3DE.","upvotes":3,"discussionId":"6a291b67e7d78ea7587e56bf","projectPage":"https://blog.gensyn.ai/look-beyond-one-size-fits-all-llms-with-ir3de/","githubRepo":"https://github.com/gensyn-ai/IR3DE","githubRepoAddedBy":"user","ai_summary":"A ridge regression-based routing method achieves competitive performance in selecting domain-expert LLMs for different tasks while enabling dynamic addition/removal of experts without retraining.","ai_keywords":["inference routers","domain-expert LLMs","ridge regression","causal language modeling","next-token prediction","reasoning tasks","linear router"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":2,"organization":{"_id":"66d2530bb26010e571f0ea9b","name":"Gensyn","fullname":"Gensyn","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/66d252ec8a438492b0d6e4ce/KD6rJavI2N74-2aT7NiZc.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"673c712c8a9ba30a57741a8c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/673c712c8a9ba30a57741a8c/khBSYmO5V1BM8Q_uPLLWB.jpeg","isPro":false,"fullname":"Oguzhan Ersoy","user":"oguzer","type":"user"},{"_id":"68e407dd07d44bad5e4e696b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/Ux50uRULEu5d_Ep4aYNNk.png","isPro":false,"fullname":"Eros Fanì","user":"Erosinho","type":"user"},{"_id":"689a0dd01db955f4e89167ce","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/CvLYwnlXHPQ_n83leKcZR.png","isPro":false,"fullname":"Semih Kara","user":"semihkara","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"66d2530bb26010e571f0ea9b","name":"Gensyn","fullname":"Gensyn","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/66d252ec8a438492b0d6e4ce/KD6rJavI2N74-2aT7NiZc.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.06098.md"}">
IR3DE: A Linear Router for Large Language Models
Abstract
A ridge regression-based routing method achieves competitive performance in selecting domain-expert LLMs for different tasks while enabling dynamic addition/removal of experts without retraining.
Foundational Large Language Models (LLMs) demonstrate proficiency on a wide range of general tasks, and achieve remarkable results on various specialized tasks via domain-expert LLMs. With the ever-growing list of available LLMs, inference routers are being proposed to select the most appropriate LLM for each prompt. However, existing routing methods either optimize cost across weak-to-strong generalist LLMs or require substantial training to support domain-expertise routing. In this paper, we propose IR3DE, a Ridge Regression-based Router for Domain Experts that provides cheap and fast routing decisions for each prompt. We evaluate IR3DE in two Causal Language Modeling (CLM) settings where the tasks are next-token prediction for all domains, and one reasoning setting where each domain has its own distinct reasoning task. Despite being a linear router, IR3DE achieves performance comparable to the other baselines in both CLM settings, and surpassing them in the reasoning setting, with a normalized performance of 98.4%. Moreover, IR3DE enables the addition or removal of new domain experts without requiring the router to be retrained from scratch, allowing a dynamic set of LLMs to be served with minimal disruption to the router itself. Our code is available at: github.com/gensyn-ai/IR3DE.
Community
IR3DE is a lightweight, modular, linear router that accurately matches each prompt to the most appropriate LLM expert.
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/2606.06098 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.06098 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.06098 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.