$\"benchmark\"$

\n","updatedAt":"2026-02-03T07:07:41.503Z","author":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","fullname":"Andong Chen","name":"AndongChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.4373926520347595},"editors":["AndongChen"],"editorAvatarUrls":["/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg"],"reactions":[{"reaction":"❤️","users":["AndongChen"],"count":1}],"isReport":false}},{"id":"6981b44c6d327ae1fb48c6ec","author":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","fullname":"Andong Chen","name":"AndongChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false},"createdAt":"2026-02-03T08:39:40.000Z","type":"comment","data":{"edited":true,"hidden":true,"hiddenBy":"","hiddenReason":"Resolved","latest":{"raw":"This comment has been hidden","html":"This comment has been hidden","updatedAt":"2026-02-03T14:01:23.557Z","author":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","fullname":"Andong Chen","name":"AndongChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"editors":[],"editorAvatarUrls":[],"reactions":[]}},{"id":"6981ff7b65a27fe4a942d39b","author":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","fullname":"Andong Chen","name":"AndongChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false},"createdAt":"2026-02-03T14:00:27.000Z","type":"comment","data":{"edited":true,"hidden":true,"hiddenBy":"","hiddenReason":"Resolved","latest":{"raw":"This comment has been hidden","html":"This comment has been hidden","updatedAt":"2026-02-03T14:01:31.572Z","author":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","fullname":"Andong Chen","name":"AndongChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"editors":[],"editorAvatarUrls":[],"reactions":[]}},{"id":"6982a3b9e9bd903ff9a95327","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":317,"isUserFollowing":false},"createdAt":"2026-02-04T01:41:13.000Z","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* [DiffThinker: Towards Generative Multimodal Reasoning with Diffusion Models](https://huggingface.co/papers/2512.24165) (2025)\n* [Thinking with Blueprints: Assisting Vision-Language Models in Spatial Reasoning via Structured Object Representation](https://huggingface.co/papers/2601.01984) (2026)\n* [InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Search](https://huggingface.co/papers/2512.18745) (2025)\n* [MM-CoT:A Benchmark for Probing Visual Chain-of-Thought Reasoning in Multimodal Models](https://huggingface.co/papers/2512.08228) (2025)\n* [Visual Generation Unlocks Human-Like Reasoning through Multimodal World Models](https://huggingface.co/papers/2601.19834) (2026)\n* [KDCM: Reducing Hallucination in LLMs through Explicit Reasoning Structures](https://huggingface.co/papers/2601.04086) (2026)\n* [Rethinking Chain-of-Thought Reasoning for Videos](https://huggingface.co/papers/2512.09616) (2025)\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":"2026-02-04T01:41:13.926Z","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":317,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7177385091781616},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[{"reaction":"🔥","users":["AndongChen"],"count":1}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.02453","authors":[{"_id":"698183aece18b18628096311","user":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","isPro":false,"fullname":"Andong Chen","user":"AndongChen","type":"user"},"name":"Andong Chen","status":"claimed_verified","statusLastChangedAt":"2026-02-03T10:03:45.069Z","hidden":false},{"_id":"698183aece18b18628096312","name":"Wenxin Zhu","hidden":false},{"_id":"698183aece18b18628096313","name":"Qiuyu Ding","hidden":false},{"_id":"698183aece18b18628096314","name":"Yuchen Song","hidden":false},{"_id":"698183aece18b18628096315","name":"Muyun Yang","hidden":false},{"_id":"698183aece18b18628096316","name":"Tiejun Zhao","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/64e8414b2c506d28d4845f84/WM4HqBzPOtmEKdOo2oRiD.png"],"publishedAt":"2026-02-02T18:43:57.000Z","submittedOnDailyAt":"2026-02-03T11:30:27.303Z","title":"Thinking with Comics: Enhancing Multimodal Reasoning through Structured Visual Storytelling","submittedOnDailyBy":{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","isPro":false,"fullname":"Andong Chen","user":"AndongChen","type":"user"},"summary":"Chain-of-Thought reasoning has driven large language models to extend from thinking with text to thinking with images and videos. However, different modalities still have clear limitations: static images struggle to represent temporal structure, while videos introduce substantial redundancy and computational cost. In this work, we propose Thinking with Comics, a visual reasoning paradigm that uses comics as a high information-density medium positioned between images and videos. Comics preserve temporal structure, embedded text, and narrative coherence while requiring significantly lower reasoning cost. We systematically study two reasoning paths based on comics and evaluate them on a range of reasoning tasks and long-context understanding tasks. Experimental results show that Thinking with Comics outperforms Thinking with Images on multi-step temporal and causal reasoning tasks, while remaining substantially more efficient than Thinking with Video. Further analysis indicates that different comic narrative structures and styles consistently affect performance across tasks, suggesting that comics serve as an effective intermediate visual representation for improving multimodal reasoning.","upvotes":35,"discussionId":"698183afce18b18628096317","projectPage":"https://thinking-with-comics.github.io/","githubRepo":"https://github.com/andongBlue/Think-with-Comics","githubRepoAddedBy":"user","ai_summary":"Thinking with Comics emerges as an effective visual reasoning approach that bridges images and videos by leveraging comic structures for improved multimodal reasoning efficiency and performance.","ai_keywords":["visual reasoning","multimodal reasoning","temporal structure","narrative coherence","long-context understanding"],"githubStars":11},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"674aee30e78dc069e0edf52e","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/dllw9lDtNWJXzKPqlODB0.png","isPro":false,"fullname":"Xin","user":"SpaceTrave1","type":"user"},{"_id":"64e8414b2c506d28d4845f84","avatarUrl":"/avatars/42c4572b13e6b76c75cb06bc29be7f61.svg","isPro":false,"fullname":"Andong Chen","user":"AndongChen","type":"user"},{"_id":"649944b142d585137ebd7dc7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/3-XWPKEyNb0m_FmaPsG6i.jpeg","isPro":false,"fullname":"Smith Waiden","user":"Coder109","type":"user"},{"_id":"69821407a3354ccc19dd5d4e","avatarUrl":"/avatars/5e21637329910383d21d3f57b6aaf1f5.svg","isPro":false,"fullname":"li","user":"IvyJU","type":"user"},{"_id":"6049d4fcf7b7068fd1ede442","avatarUrl":"/avatars/2a44beb3c5f090d52a233e1515592902.svg","isPro":false,"fullname":"And","user":"AnTree","type":"user"},{"_id":"684d57f26e04c265777ead3f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/cuOj-bQqukSZreXgUJlfm.png","isPro":false,"fullname":"Joakim Lee","user":"Reinforcement4All","type":"user"},{"_id":"64f198aab45e86c2be566c92","avatarUrl":"/avatars/2a632503087b65bfd6d81af5ce07e21a.svg","isPro":false,"fullname":"yudi zhang","user":"YudiZh","type":"user"},{"_id":"6645ab01621ded608be13983","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6645ab01621ded608be13983/qJYix9bU6bLqBsp_bEF31.jpeg","isPro":false,"fullname":"yxdu","user":"yxdu","type":"user"},{"_id":"68a7e6b0b655f5bfee4056c1","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/kIS7HUyhzepvUn4Vy4i2j.png","isPro":false,"fullname":"Wenhao Jiang","user":"whjiang93","type":"user"},{"_id":"667c25280c384702a8b5cf6e","avatarUrl":"/avatars/2ca1298a9d7163419b6aae0b21df9cc7.svg","isPro":false,"fullname":"Wang","user":"wangwang2024","type":"user"},{"_id":"674ac97729a3bb873fc995c6","avatarUrl":"/avatars/cd5dc0bb367b552eeaefee4343adb89b.svg","isPro":false,"fullname":"Zhou Chulun","user":"Chow1997-CUHK","type":"user"},{"_id":"62d4bf8c97ab9eb08762a975","avatarUrl":"/avatars/73c6228e317cf37b4e3c3e7a4b3d8ae8.svg","isPro":false,"fullname":"Minghao Wu","user":"minghaowu","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2602.02453

Thinking with Comics: Enhancing Multimodal Reasoning through Structured Visual Storytelling

Published on Feb 2

· Submitted by

Andong Chen on Feb 3

Upvote

Authors:

Andong Chen ,

Abstract

Thinking with Comics emerges as an effective visual reasoning approach that bridges images and videos by leveraging comic structures for improved multimodal reasoning efficiency and performance.

AI-generated summary

Chain-of-Thought reasoning has driven large language models to extend from thinking with text to thinking with images and videos. However, different modalities still have clear limitations: static images struggle to represent temporal structure, while videos introduce substantial redundancy and computational cost. In this work, we propose Thinking with Comics, a visual reasoning paradigm that uses comics as a high information-density medium positioned between images and videos. Comics preserve temporal structure, embedded text, and narrative coherence while requiring significantly lower reasoning cost. We systematically study two reasoning paths based on comics and evaluate them on a range of reasoning tasks and long-context understanding tasks. Experimental results show that Thinking with Comics outperforms Thinking with Images on multi-step temporal and causal reasoning tasks, while remaining substantially more efficient than Thinking with Video. Further analysis indicates that different comic narrative structures and styles consistently affect performance across tasks, suggesting that comics serve as an effective intermediate visual representation for improving multimodal reasoning.