@librarian-bot\n\t recommend

\n","updatedAt":"2025-04-02T09:27:09.503Z","author":{"_id":"60107b385ac3e86b3ea4fc34","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg","fullname":"Daniel van Strien","name":"davanstrien","type":"user","isPro":true,"isHf":true,"isHfAdmin":false,"isMod":false,"followerCount":838,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7918877601623535},"editors":["davanstrien"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg"],"reactions":[],"isReport":false},"replies":[{"id":"67ed02f40a7438091a5b0311","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":"2025-04-02T09:27:16.000Z","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* [VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data](https://huggingface.co/papers/2502.06737) (2025)\n* [Expanding RL with Verifiable Rewards Across Diverse Domains](https://huggingface.co/papers/2503.23829) (2025)\n* [Small Models Struggle to Learn from Strong Reasoners](https://huggingface.co/papers/2502.12143) (2025)\n* [Fino1: On the Transferability of Reasoning Enhanced LLMs to Finance](https://huggingface.co/papers/2502.08127) (2025)\n* [InfiR : Crafting Effective Small Language Models and Multimodal Small Language Models in Reasoning](https://huggingface.co/papers/2502.11573) (2025)\n* [Enhancing LLM Reasoning with Iterative DPO: A Comprehensive Empirical Investigation](https://huggingface.co/papers/2503.12854) (2025)\n* [DeepThink: Aligning Language Models with Domain-Specific User Intents](https://huggingface.co/papers/2502.05497) (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":"2025-04-02T09:27:16.322Z","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":0.7357696294784546},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[{"reaction":"👍","users":["davanstrien"],"count":1}],"isReport":false,"parentCommentId":"67ed02ed8d8a8d5443d5b696"}}]}],"primaryEmailConfirmed":false,"paper":{"id":"2502.13124","authors":[{"_id":"67b5410d11c9f80e23dcc78c","name":"Weizhe Yuan","hidden":false},{"_id":"67b5410d11c9f80e23dcc78d","name":"Jane Yu","hidden":false},{"_id":"67b5410d11c9f80e23dcc78e","name":"Song Jiang","hidden":false},{"_id":"67b5410d11c9f80e23dcc78f","user":{"_id":"64c01530bd4b58768d8b46b5","avatarUrl":"/avatars/0f3c0cdab16e3078b37569d73ca03792.svg","isPro":false,"fullname":"Karthik padthe","user":"Karthikpadthe","type":"user"},"name":"Karthik Padthe","status":"claimed_verified","statusLastChangedAt":"2025-12-08T08:37:24.697Z","hidden":false},{"_id":"67b5410d11c9f80e23dcc790","name":"Yang Li","hidden":false},{"_id":"67b5410d11c9f80e23dcc791","user":{"_id":"640e26b03830fd441c2c106a","avatarUrl":"/avatars/8e457552ec454cbaa3ed9e138b29b47a.svg","isPro":false,"fullname":"Dong Wang","user":"dongwang218","type":"user"},"name":"Dong Wang","status":"claimed_verified","statusLastChangedAt":"2025-08-06T19:40:21.887Z","hidden":false},{"_id":"67b5410d11c9f80e23dcc792","name":"Ilia Kulikov","hidden":false},{"_id":"67b5410d11c9f80e23dcc793","name":"Kyunghyun Cho","hidden":false},{"_id":"67b5410d11c9f80e23dcc794","name":"Yuandong Tian","hidden":false},{"_id":"67b5410d11c9f80e23dcc795","name":"Jason E Weston","hidden":false},{"_id":"67b5410d11c9f80e23dcc796","user":{"_id":"659a395421a7431643caedda","avatarUrl":"/avatars/c1e0bbcedce68fe3b4fe39e0cf01c65c.svg","isPro":false,"fullname":"Xian Li","user":"xlxxl","type":"user"},"name":"Xian Li","status":"extracted_confirmed","statusLastChangedAt":"2025-05-22T23:46:00.937Z","hidden":false}],"publishedAt":"2025-02-18T18:46:57.000Z","title":"NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions","summary":"Scaling reasoning capabilities beyond traditional domains such as math and\ncoding is hindered by the lack of diverse and high-quality questions. To\novercome this limitation, we introduce a scalable approach for generating\ndiverse and challenging reasoning questions, accompanied by reference answers.\nWe present NaturalReasoning, a comprehensive dataset comprising 2.8 million\nquestions that span multiple domains, including STEM fields (e.g., Physics,\nComputer Science), Economics, Social Sciences, and more. We demonstrate the\nutility of the questions in NaturalReasoning through knowledge distillation\nexperiments which show that NaturalReasoning can effectively elicit and\ntransfer reasoning capabilities from a strong teacher model. Furthermore, we\ndemonstrate that NaturalReasoning is also effective for unsupervised\nself-training using external reward models or self-rewarding.","upvotes":7,"discussionId":"67b5410f11c9f80e23dcc861","ai_summary":"NaturalReasoning, a dataset of 2.8 million diverse reasoning questions, enhances knowledge distillation and self-training across various domains, including STEM, Economics, and Social Sciences.","ai_keywords":["natural reasoning","knowledge distillation","self-training","external reward models","self-rewarding"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64bbe9b236eb058cd9d6a5b9","avatarUrl":"/avatars/c7c01a3fa8809e73800392679abff6d5.svg","isPro":false,"fullname":"Kai Zuberbühler","user":"kaizuberbuehler","type":"user"},{"_id":"63b29f54922f26a27e72fbb4","avatarUrl":"/avatars/c36a08d3aacf953533b89d7dcfa5cf18.svg","isPro":false,"fullname":"Alexander","user":"alex43219","type":"user"},{"_id":"645523ce1a543cf97b1dbdcd","avatarUrl":"/avatars/67af5c6846e4ddfdefa2bb344336185d.svg","isPro":false,"fullname":"Tim Dingman","user":"tdingman","type":"user"},{"_id":"63f0760ef1a47aaea5be1e6b","avatarUrl":"/avatars/795178a2d7f92150a6d1796f288c2f05.svg","isPro":false,"fullname":"Ji-Xiang","user":"Ji-Xiang","type":"user"},{"_id":"642a671569bb3e94e9a5923e","avatarUrl":"/avatars/dd64325e910368acf98f9e384a6850cc.svg","isPro":false,"fullname":"Xiang Fu","user":"craigxiangfu","type":"user"},{"_id":"60107b385ac3e86b3ea4fc34","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg","isPro":true,"fullname":"Daniel van Strien","user":"davanstrien","type":"user"},{"_id":"63c1699e40a26dd2db32400d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63c1699e40a26dd2db32400d/3N0-Zp8igv8-52mXAdiiq.jpeg","isPro":false,"fullname":"Chroma","user":"Chroma111","type":"user"}],"acceptLanguages":["*"]}">

Papers

arxiv:2502.13124

NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions

Published on Feb 18, 2025

Upvote

Authors:

Karthik Padthe ,

Dong Wang ,

Xian Li

Abstract

NaturalReasoning, a dataset of 2.8 million diverse reasoning questions, enhances knowledge distillation and self-training across various domains, including STEM, Economics, and Social Sciences.

AI-generated summary

Scaling reasoning capabilities beyond traditional domains such as math and coding is hindered by the lack of diverse and high-quality questions. To overcome this limitation, we introduce a scalable approach for generating diverse and challenging reasoning questions, accompanied by reference answers. We present NaturalReasoning, a comprehensive dataset comprising 2.8 million questions that span multiple domains, including STEM fields (e.g., Physics, Computer Science), Economics, Social Sciences, and more. We demonstrate the utility of the questions in NaturalReasoning through knowledge distillation experiments which show that NaturalReasoning can effectively elicit and transfer reasoning capabilities from a strong teacher model. Furthermore, we demonstrate that NaturalReasoning is also effective for unsupervised self-training using external reward models or self-rewarding.

View arXiv page View PDF Add to collection