FAAST compiles labeled examples into fast weights in a single forward pass, eliminating memory or context dependencies and achieving constant-time inference.</p>\n","updatedAt":"2026-05-14T12:47:50.745Z","author":{"_id":"645dbf365ebf379fd6dc7119","avatarUrl":"/avatars/5f946cea8fa8633b7091e3e3c5258906.svg","fullname":"Guangsheng Bao","name":"gshbao","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7982141971588135},"editors":["gshbao"],"editorAvatarUrls":["/avatars/5f946cea8fa8633b7091e3e3c5258906.svg"],"reactions":[],"isReport":false}},{"id":"6a067a2805c94743d06866de","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":355,"isUserFollowing":false},"createdAt":"2026-05-15T01:43:04.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [Test-Time Instance-Specific Parameter Composition: A New Paradigm for Adaptive Generative Modeling](https://huggingface.co/papers/2603.27665) (2026)\n* [Learning When to Attend: Conditional Memory Access for Long-Context LLMs](https://huggingface.co/papers/2603.17484) (2026)\n* [OASIS: Online Activation Subspace Learning for Memory-Efficient Training](https://huggingface.co/papers/2604.09406) (2026)\n* [Mixture of Chapters: Scaling Learnt Memory in Transformers](https://huggingface.co/papers/2603.21096) (2026)\n* [Training-Free Test-Time Contrastive Learning for Large Language Models](https://huggingface.co/papers/2604.13552) (2026)\n* [Cross-Modal Bayesian Low-Rank Adaptation for Uncertainty-Aware Multimodal Learning](https://huggingface.co/papers/2604.16657) (2026)\n* [Improving Sparse Memory Finetuning](https://huggingface.co/papers/2604.05248) (2026)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"<p>This is an automated message from the <a href=\"https://huggingface.co/librarian-bots\">Librarian Bot</a>. I found the following papers similar to this paper. </p>\n<p>The following papers were recommended by the Semantic Scholar API </p>\n<ul>\n<li><a href=\"https://huggingface.co/papers/2603.27665\">Test-Time Instance-Specific Parameter Composition: A New Paradigm for Adaptive Generative Modeling</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2603.17484\">Learning When to Attend: Conditional Memory Access for Long-Context LLMs</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.09406\">OASIS: Online Activation Subspace Learning for Memory-Efficient Training</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2603.21096\">Mixture of Chapters: Scaling Learnt Memory in Transformers</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.13552\">Training-Free Test-Time Contrastive Learning for Large Language Models</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.16657\">Cross-Modal Bayesian Low-Rank Adaptation for Uncertainty-Aware Multimodal Learning</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.05248\">Improving Sparse Memory Finetuning</a> (2026)</li>\n</ul>\n<p> Please give a thumbs up to this comment if you found it helpful!</p>\n<p> If you want recommendations for any Paper on Hugging Face checkout <a href=\"https://huggingface.co/spaces/librarian-bots/recommend_similar_papers\">this</a> Space</p>\n<p> You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: <code><span class=\"SVELTE_PARTIAL_HYDRATER contents\" data-target=\"UserMention\" data-props=\"{&quot;user&quot;:&quot;librarian-bot&quot;}\"><span class=\"inline-block\"><span class=\"contents\"><a href=\"/librarian-bot\">@<span class=\"underline\">librarian-bot</span></a></span> </span></span> recommend</code></p>\n","updatedAt":"2026-05-15T01:43:04.964Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":355,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.688958466053009},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.04651","authors":[{"_id":"6a059c55b1a8cbabc9f089e2","user":{"_id":"645dbf365ebf379fd6dc7119","avatarUrl":"/avatars/5f946cea8fa8633b7091e3e3c5258906.svg","isPro":false,"fullname":"Guangsheng Bao","user":"gshbao","type":"user","name":"gshbao"},"name":"Guangsheng Bao","status":"claimed_verified","statusLastChangedAt":"2026-05-14T10:54:07.174Z","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e3","name":"Hongbo Zhang","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e4","name":"Han Cui","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e5","name":"Ke Sun","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e6","name":"Yanbin Zhao","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e7","name":"Juncai He","hidden":false},{"_id":"6a059c55b1a8cbabc9f089e8","name":"Yue Zhang","hidden":false}],"publishedAt":"2026-05-08T00:00:00.000Z","submittedOnDailyAt":"2026-05-14T00:00:00.000Z","title":"FAAST: Forward-Only Associative Learning via Closed-Form Fast Weights for Test-Time Supervised Adaptation","submittedOnDailyBy":{"_id":"645dbf365ebf379fd6dc7119","avatarUrl":"/avatars/5f946cea8fa8633b7091e3e3c5258906.svg","isPro":false,"fullname":"Guangsheng Bao","user":"gshbao","type":"user","name":"gshbao"},"summary":"Adapting pretrained models typically involves a trade-off between the high training costs of backpropagation and the heavy inference overhead of memory-based or in-context learning. We propose FAAST, a forward-only associative adaptation method that analytically compiles labeled examples into fast weights in a single pass. By eliminating memory or context dependence, FAAST achieves constant-time inference and decouples task adaptation from pretrained representation. Across image classification and language modeling benchmarks, FAAST matches or exceeds backprop-based adaptation while reducing adaptation time by over 90% and is competitive to memory/context-based adaptation while saving memory usage by up to 95%. These results demonstrate FAAST as a highly efficient, scalable solution for supervised task adaptation, particularly for resource-constrained models. We release the code and models at https://github.com/baoguangsheng/faast.","upvotes":1,"discussionId":"6a059c55b1a8cbabc9f089e9","githubRepo":"https://github.com/baoguangsheng/faast","githubRepoAddedBy":"user","ai_summary":"FAAST enables efficient task adaptation by compiling labeled examples into fast weights through forward-only computation, achieving significant speedup and memory savings over traditional backpropagation methods.","ai_keywords":["pretrained models","backpropagation","memory-based learning","in-context learning","forward-only associative adaptation","fast weights","task adaptation","supervised task adaptation"],"githubStars":3,"organization":{"_id":"66bb231e40d36c70d6ad0c4b","name":"WestlakeNLP","fullname":"Text Intelligence Lab of Westlake University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/622ee9f3165ba2c1bcbc7706/KpIm3isRczYp7kSnfNGSL.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"69830919a33f96b73181cb00","avatarUrl":"/avatars/d89a054a911b5552af13ca1be5a8270d.svg","isPro":false,"fullname":"Luca Moretti","user":"based-user","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"66bb231e40d36c70d6ad0c4b","name":"WestlakeNLP","fullname":"Text Intelligence Lab of Westlake University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/622ee9f3165ba2c1bcbc7706/KpIm3isRczYp7kSnfNGSL.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.04651.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2605.04651 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2605.04651 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.