here

\n","updatedAt":"2024-11-22T09:00:18.328Z","author":{"_id":"6374c494958cd71fa7ea0a9d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6374c494958cd71fa7ea0a9d/b2SjfvbjYqPCW38LzkzWl.jpeg","fullname":"yuyijiong","name":"yuyijiong","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":52,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.9359161257743835},"editors":["yuyijiong"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6374c494958cd71fa7ea0a9d/b2SjfvbjYqPCW38LzkzWl.jpeg"],"reactions":[],"isReport":false}},{"id":"674050b0bed7c8432375df68","author":{"_id":"6323f399462470712720c155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6323f399462470712720c155/SWsMNa7vETUSrOt9Qf-oe.png","fullname":"Yinxu Pan","name":"cppowboy","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":15,"isUserFollowing":false},"createdAt":"2024-11-22T09:36:48.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Patience is the key in life.","html":"

Patience is the key in life.

\n","updatedAt":"2024-11-22T09:36:48.942Z","author":{"_id":"6323f399462470712720c155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6323f399462470712720c155/SWsMNa7vETUSrOt9Qf-oe.png","fullname":"Yinxu Pan","name":"cppowboy","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":15,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.988767683506012},"editors":["cppowboy"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6323f399462470712720c155/SWsMNa7vETUSrOt9Qf-oe.png"],"reactions":[],"isReport":false}},{"id":"6741311a26c0539fc0351897","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":"2024-11-23T01:34:18.000Z","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* [SuperCorrect: Supervising and Correcting Language Models with Error-Driven Insights](https://huggingface.co/papers/2410.09008) (2024)\n* [PersonaMath: Enhancing Math Reasoning through Persona-Driven Data Augmentation](https://huggingface.co/papers/2410.01504) (2024)\n* [AtomThink: A Slow Thinking Framework for Multimodal Mathematical Reasoning](https://huggingface.co/papers/2411.11930) (2024)\n* [A Comparative Study on Reasoning Patterns of OpenAI's o1 Model](https://huggingface.co/papers/2410.13639) (2024)\n* [Vision-Language Models Can Self-Improve Reasoning via Reflection](https://huggingface.co/papers/2411.00855) (2024)\n* [BEATS: Optimizing LLM Mathematical Capabilities with BackVerify and Adaptive Disambiguate based Efficient Tree Search](https://huggingface.co/papers/2409.17972) (2024)\n* [Step Guided Reasoning: Improving Mathematical Reasoning using Guidance Generation and Step Reasoning](https://huggingface.co/papers/2410.19817) (2024)\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":"2024-11-23T01:34:18.145Z","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":0.7187904119491577},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2411.13082","authors":[{"_id":"673ebf5a4edfc6d3b57b8376","user":{"_id":"6374c494958cd71fa7ea0a9d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6374c494958cd71fa7ea0a9d/b2SjfvbjYqPCW38LzkzWl.jpeg","isPro":false,"fullname":"yuyijiong","user":"yuyijiong","type":"user"},"name":"Yijiong Yu","status":"admin_assigned","statusLastChangedAt":"2024-11-22T13:21:54.207Z","hidden":false}],"publishedAt":"2024-11-20T07:20:48.000Z","submittedOnDailyAt":"2024-11-22T06:29:15.009Z","title":"Patience Is The Key to Large Language Model Reasoning","submittedOnDailyBy":{"_id":"6374c494958cd71fa7ea0a9d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6374c494958cd71fa7ea0a9d/b2SjfvbjYqPCW38LzkzWl.jpeg","isPro":false,"fullname":"yuyijiong","user":"yuyijiong","type":"user"},"summary":"Recent advancements in the field of large language models, particularly\nthrough the Chain of Thought (CoT) approach, have demonstrated significant\nimprovements in solving complex problems. However, existing models either tend\nto sacrifice detailed reasoning for brevity due to user preferences, or require\nextensive and expensive training data to learn complicated reasoning ability,\nlimiting their potential in solving complex tasks. To bridge this gap,\nfollowing the concept of scaling test-time, we propose a simple method by\nencouraging models to adopt a more patient reasoning style without the need of\nintroducing new knowledge or skills. To employ a preference optimization\napproach, we generate detailed reasoning processes as positive examples and\nsimple answers as negative examples, thereby training the model to favor\nthoroughness in its responses. Our results demonstrate a performance increase\nof up to 6.7% on GSM8k with training just on a lightweight dataset.","upvotes":7,"discussionId":"673ebf5b4edfc6d3b57b83a7","ai_summary":"A method encouraging detailed reasoning in large language models via preference optimization improves performance on complex tasks using a lightweight dataset.","ai_keywords":["Chain of Thought","large language models","patient reasoning","preference optimization","GSM8k"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6374c494958cd71fa7ea0a9d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6374c494958cd71fa7ea0a9d/b2SjfvbjYqPCW38LzkzWl.jpeg","isPro":false,"fullname":"yuyijiong","user":"yuyijiong","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"643b19f8a856622f978df30f","avatarUrl":"/avatars/c82779fdf94f80cdb5020504f83c818b.svg","isPro":false,"fullname":"Yatharth Sharma","user":"YaTharThShaRma999","type":"user"},{"_id":"64d4615cf8082bf19b916492","avatarUrl":"/avatars/8e1b59565ec5e4b31090cf1b911781b9.svg","isPro":false,"fullname":"wongyukim","user":"wongyukim","type":"user"},{"_id":"6394a030a1be57c121dca76d","avatarUrl":"/avatars/eb58592659edcda44ce5658964cd1ad0.svg","isPro":false,"fullname":"Bur","user":"KSa","type":"user"},{"_id":"6560d75d6ff1b91e28e3cd7b","avatarUrl":"/avatars/bf205b47c71b197c56414ad1aaae3453.svg","isPro":false,"fullname":"js","user":"rldy","type":"user"},{"_id":"648d2e2e514bf0ce32ba729f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/648d2e2e514bf0ce32ba729f/VPL1rehLxkvixz5oRD6u_.jpeg","isPro":false,"fullname":"Yaolun Zhang","user":"Mercury7353","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2411.13082

Patience Is The Key to Large Language Model Reasoning

Published on Nov 20, 2024

· Submitted by

yuyijiong on Nov 22, 2024

Upvote

Authors:

Yijiong Yu

Abstract

A method encouraging detailed reasoning in large language models via preference optimization improves performance on complex tasks using a lightweight dataset.

AI-generated summary

Recent advancements in the field of large language models, particularly through the Chain of Thought (CoT) approach, have demonstrated significant improvements in solving complex problems. However, existing models either tend to sacrifice detailed reasoning for brevity due to user preferences, or require extensive and expensive training data to learn complicated reasoning ability, limiting their potential in solving complex tasks. To bridge this gap, following the concept of scaling test-time, we propose a simple method by encouraging models to adopt a more patient reasoning style without the need of introducing new knowledge or skills. To employ a preference optimization approach, we generate detailed reasoning processes as positive examples and simple answers as negative examples, thereby training the model to favor thoroughness in its responses. Our results demonstrate a performance increase of up to 6.7% on GSM8k with training just on a lightweight dataset.

View arXiv page View PDF Add to collection