https://jasper0314-huang.github.io/fast-thinkact/

\n","updatedAt":"2026-01-15T02:40:27.539Z","author":{"_id":"64705d224be5cf1f3348d6bc","avatarUrl":"/avatars/270bff7c7cb326528dc192fc38561a8b.svg","fullname":"Chi-Pin Huang","name":"jasper0314-huang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.3701716959476471},"editors":["jasper0314-huang"],"editorAvatarUrls":["/avatars/270bff7c7cb326528dc192fc38561a8b.svg"],"reactions":[],"isReport":false}},{"id":"6969962c069de7a387b2cd0f","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":318,"isUserFollowing":false},"createdAt":"2026-01-16T01:36:44.000Z","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* [VLingNav: Embodied Navigation with Adaptive Reasoning and Visual-Assisted Linguistic Memory](https://huggingface.co/papers/2601.08665) (2026)\n* [Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training](https://huggingface.co/papers/2512.24125) (2025)\n* [Unifying Perception and Action: A Hybrid-Modality Pipeline with Implicit Visual Chain-of-Thought for Robotic Action Generation](https://huggingface.co/papers/2511.19859) (2025)\n* [MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots](https://huggingface.co/papers/2511.17889) (2025)\n* [CoT4AD: A Vision-Language-Action Model with Explicit Chain-of-Thought Reasoning for Autonomous Driving](https://huggingface.co/papers/2511.22532) (2025)\n* [DualVLA: Building a Generalizable Embodied Agent via Partial Decoupling of Reasoning and Action](https://huggingface.co/papers/2511.22134) (2025)\n* [LaST$_{0}$: Latent Spatio-Temporal Chain-of-Thought for Robotic Vision-Language-Action Model](https://huggingface.co/papers/2601.05248) (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":"

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2026-01-16T01:36:44.641Z","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":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7236086130142212},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2601.09708","authors":[{"_id":"69684f740ac10a06522f69ba","user":{"_id":"64705d224be5cf1f3348d6bc","avatarUrl":"/avatars/270bff7c7cb326528dc192fc38561a8b.svg","isPro":false,"fullname":"Chi-Pin Huang","user":"jasper0314-huang","type":"user"},"name":"Chi-Pin Huang","status":"claimed_verified","statusLastChangedAt":"2026-01-15T15:03:32.512Z","hidden":false},{"_id":"69684f740ac10a06522f69bb","name":"Yunze Man","hidden":false},{"_id":"69684f740ac10a06522f69bc","name":"Zhiding Yu","hidden":false},{"_id":"69684f740ac10a06522f69bd","user":{"_id":"64ae22dd1aee69ece065cdcd","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ae22dd1aee69ece065cdcd/JG7QaHIrr4i2k4uwR4pZK.png","isPro":false,"fullname":"Min-Hung Chen","user":"cmhungsteve","type":"user"},"name":"Min-Hung Chen","status":"claimed_verified","statusLastChangedAt":"2026-01-15T15:03:27.724Z","hidden":false},{"_id":"69684f740ac10a06522f69be","name":"Jan Kautz","hidden":false},{"_id":"69684f740ac10a06522f69bf","name":"Yu-Chiang Frank Wang","hidden":false},{"_id":"69684f740ac10a06522f69c0","user":{"_id":"6312cab05beb528b5c1500e3","avatarUrl":"/avatars/a328e8cc99fb031b2d5c911c4b577e7e.svg","isPro":false,"fullname":"Fu-En Yang","user":"FuEnYang","type":"user"},"name":"Fu-En Yang","status":"claimed_verified","statusLastChangedAt":"2026-01-15T15:03:30.478Z","hidden":false}],"publishedAt":"2026-01-14T18:59:59.000Z","submittedOnDailyAt":"2026-01-15T00:10:27.528Z","title":"Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning","submittedOnDailyBy":{"_id":"64705d224be5cf1f3348d6bc","avatarUrl":"/avatars/270bff7c7cb326528dc192fc38561a8b.svg","isPro":false,"fullname":"Chi-Pin Huang","user":"jasper0314-huang","type":"user"},"summary":"Vision-Language-Action (VLA) tasks require reasoning over complex visual scenes and executing adaptive actions in dynamic environments. While recent studies on reasoning VLAs show that explicit chain-of-thought (CoT) can improve generalization, they suffer from high inference latency due to lengthy reasoning traces. We propose Fast-ThinkAct, an efficient reasoning framework that achieves compact yet performant planning through verbalizable latent reasoning. Fast-ThinkAct learns to reason efficiently with latent CoTs by distilling from a teacher, driven by a preference-guided objective to align manipulation trajectories that transfers both linguistic and visual planning capabilities for embodied control. This enables reasoning-enhanced policy learning that effectively connects compact reasoning to action execution. Extensive experiments across diverse embodied manipulation and reasoning benchmarks demonstrate that Fast-ThinkAct achieves strong performance with up to 89.3\\% reduced inference latency over state-of-the-art reasoning VLAs, while maintaining effective long-horizon planning, few-shot adaptation, and failure recovery.","upvotes":53,"discussionId":"69684f740ac10a06522f69c1","projectPage":"https://jasper0314-huang.github.io/fast-thinkact/","ai_summary":"Fast-ThinkAct is an efficient vision-language-action framework that reduces inference latency by 89.3% through compact latent reasoning while maintaining long-horizon planning and few-shot adaptation capabilities.","ai_keywords":["chain-of-thought","latent reasoning","preference-guided objective","embodied control","policy learning","inference latency","vision-language-action"],"organization":{"_id":"60262b67268c201cdc8b7d43","name":"nvidia","fullname":"NVIDIA","avatar":"https://cdn-uploads.huggingface.co/production/uploads/1613114437487-60262a8e0703121c822a80b6.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64705d224be5cf1f3348d6bc","avatarUrl":"/avatars/270bff7c7cb326528dc192fc38561a8b.svg","isPro":false,"fullname":"Chi-Pin Huang","user":"jasper0314-huang","type":"user"},{"_id":"6312cab05beb528b5c1500e3","avatarUrl":"/avatars/a328e8cc99fb031b2d5c911c4b577e7e.svg","isPro":false,"fullname":"Fu-En Yang","user":"FuEnYang","type":"user"},{"_id":"65d070536130ef7be09b92d0","avatarUrl":"/avatars/abb2a4d1d3376ec46c6e2b9105232623.svg","isPro":false,"fullname":"Yung-Hsuan Lai","user":"NTUBarista","type":"user"},{"_id":"66c8037c737ba92ae3fe0322","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66c8037c737ba92ae3fe0322/WR_Yh5DWOVVh7IFlF24NM.jpeg","isPro":true,"fullname":"Zhiding Yu","user":"Zhiding","type":"user"},{"_id":"66631978126013dc6a1fe62d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66631978126013dc6a1fe62d/PK0ZA6ic_iMuC230ZzWeh.png","isPro":false,"fullname":"Yu-Cheng Chou","user":"Johnson111788","type":"user"},{"_id":"61703fa3dff0ef663e421ab5","avatarUrl":"/avatars/96172e2782e218bbbddfdf47f96c1ad4.svg","isPro":false,"fullname":"Jaehun Jung","user":"Jaehun","type":"user"},{"_id":"6463554dd2044cd1d7c6e0bf","avatarUrl":"/avatars/d7653623117268c545a7063fec69664b.svg","isPro":false,"fullname":"Bingzheng Wei","user":"Bingzheng","type":"user"},{"_id":"66e7dceb0d8a243f944887a7","avatarUrl":"/avatars/07d7d90ab58890bbc0d3e82350be33fb.svg","isPro":false,"fullname":"Chou","user":"Zi-Ting","type":"user"},{"_id":"65dad90a80bafdfb4bc94dc4","avatarUrl":"/avatars/7131b82e11245744760e8a45ba914161.svg","isPro":false,"fullname":"Kai-Po Chang","user":"kaipochang0810","type":"user"},{"_id":"68dd25dfa9994b9d8761a287","avatarUrl":"/avatars/dc799f4bafd59b95fea09d008383d254.svg","isPro":false,"fullname":"Kervin","user":"qervin","type":"user"},{"_id":"64ae22dd1aee69ece065cdcd","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ae22dd1aee69ece065cdcd/JG7QaHIrr4i2k4uwR4pZK.png","isPro":false,"fullname":"Min-Hung Chen","user":"cmhungsteve","type":"user"},{"_id":"686f806aeb53e7adba46c3de","avatarUrl":"/avatars/db11c84d602cefa72ba409c8292e4191.svg","isPro":true,"fullname":"guoguoc","user":"woshichaoren123","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0,"organization":{"_id":"60262b67268c201cdc8b7d43","name":"nvidia","fullname":"NVIDIA","avatar":"https://cdn-uploads.huggingface.co/production/uploads/1613114437487-60262a8e0703121c822a80b6.png"}}">

Papers

arxiv:2601.09708

Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning

Published on Jan 14

· Submitted by

Chi-Pin Huang on Jan 15

NVIDIA

Upvote

Authors:

Chi-Pin Huang ,

Min-Hung Chen ,

Fu-En Yang

Abstract

Fast-ThinkAct is an efficient vision-language-action framework that reduces inference latency by 89.3% through compact latent reasoning while maintaining long-horizon planning and few-shot adaptation capabilities.

AI-generated summary

Vision-Language-Action (VLA) tasks require reasoning over complex visual scenes and executing adaptive actions in dynamic environments. While recent studies on reasoning VLAs show that explicit chain-of-thought (CoT) can improve generalization, they suffer from high inference latency due to lengthy reasoning traces. We propose Fast-ThinkAct, an efficient reasoning framework that achieves compact yet performant planning through verbalizable latent reasoning. Fast-ThinkAct learns to reason efficiently with latent CoTs by distilling from a teacher, driven by a preference-guided objective to align manipulation trajectories that transfers both linguistic and visual planning capabilities for embodied control. This enables reasoning-enhanced policy learning that effectively connects compact reasoning to action execution. Extensive experiments across diverse embodied manipulation and reasoning benchmarks demonstrate that Fast-ThinkAct achieves strong performance with up to 89.3\% reduced inference latency over state-of-the-art reasoning VLAs, while maintaining effective long-horizon planning, few-shot adaptation, and failure recovery.