Librarian Bot. I found the following papers similar to this paper.

\n

The following papers were recommended by the Semantic Scholar API

\n

\n
• DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence (2024)
\n
• ReFT: Reasoning with Reinforced Fine-Tuning (2024)
\n
• SciGLM: Training Scientific Language Models with Self-Reflective Instruction Annotation and Tuning (2024)
\n
• DeepSeek LLM: Scaling Open-Source Language Models with Longtermism (2024)
\n
• Augmenting Math Word Problems via Iterative Question Composing (2024)
\n

\n

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

\n

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

\n

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

\n","updatedAt":"2024-02-07T01:21:51.880Z","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":264}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7435516715049744},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}},{"id":"65c72687469efddc1ccea585","author":{"_id":"64bd167fb7375f6b8460f52d","avatarUrl":"/avatars/2036be85499a9d3002d38416a7cfaa31.svg","fullname":"zaid zameer shaikh","name":"zaidzameer010","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false},"createdAt":"2024-02-10T07:32:23.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"deepseek ain't no joke","html":"

deepseek ain't no joke

\n","updatedAt":"2024-02-10T07:32:42.741Z","author":{"_id":"64bd167fb7375f6b8460f52d","avatarUrl":"/avatars/2036be85499a9d3002d38416a7cfaa31.svg","fullname":"zaid zameer shaikh","name":"zaidzameer010","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.2873454689979553},"editors":["zaidzameer010"],"editorAvatarUrls":["/avatars/2036be85499a9d3002d38416a7cfaa31.svg"],"reactions":[{"reaction":"👍","users":["DINESHKUMARM","wjmcat"],"count":2}],"isReport":false}},{"id":"664e9268f79b50be1443ab35","author":{"_id":"664422b5044a16758a21cc7f","avatarUrl":"/avatars/5701da5bb1a7189d2230350688530401.svg","fullname":"Yang","name":"fai1165","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false},"createdAt":"2024-05-23T00:48:40.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"So can we use GRPO to tune LLM models now? \nI mean is GRPO supported and open-source on Huggingface?","html":"

So can we use GRPO to tune LLM models now?
I mean is GRPO supported and open-source on Huggingface?

\n","updatedAt":"2024-05-23T00:48:40.591Z","author":{"_id":"664422b5044a16758a21cc7f","avatarUrl":"/avatars/5701da5bb1a7189d2230350688530401.svg","fullname":"Yang","name":"fai1165","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9416334629058838},"editors":["fai1165"],"editorAvatarUrls":["/avatars/5701da5bb1a7189d2230350688530401.svg"],"reactions":[{"reaction":"👀","users":["yiran-wang3","diegomrodrigues","SugarSkullFace","wjmcat","ajcs","sumching","catmimi","Yyy0530"],"count":8}],"isReport":false}},{"id":"66503992729c53d860a54451","author":{"_id":"64f81458ec913b4f974c43d6","avatarUrl":"/avatars/5498f5cc79976d9e60ea8fff3a9a68b3.svg","fullname":"One","name":"Onelaststeptojoin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false},"createdAt":"2024-05-24T06:54:10.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"1+1\n","html":"

1+1

\n","updatedAt":"2024-05-24T06:54:10.822Z","author":{"_id":"64f81458ec913b4f974c43d6","avatarUrl":"/avatars/5498f5cc79976d9e60ea8fff3a9a68b3.svg","fullname":"One","name":"Onelaststeptojoin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"ar","probability":0.3478406071662903},"editors":["Onelaststeptojoin"],"editorAvatarUrls":["/avatars/5498f5cc79976d9e60ea8fff3a9a68b3.svg"],"reactions":[],"isReport":false}},{"id":"6665096ec9dd54387c865288","author":{"_id":"6186ddf6a7717cb375090c01","avatarUrl":"/avatars/716b6a7d1094c8036b2a8a7b9063e8aa.svg","fullname":"Julien BLANCHON","name":"blanchon","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":142},"createdAt":"2024-06-09T01:46:22.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"# DeepSeekMath: Revolutionizing Mathematical Reasoning in Open-Source AI\n\nhttps://cdn-uploads.huggingface.co/production/uploads/6186ddf6a7717cb375090c01/K82WigafgoM6WxaqHk4ns.mp4 \n\n## Links 🔗:\n👉 Subscribe: https://www.youtube.com/@Arxflix\n👉 Twitter: https://x.com/arxflix\n👉 LMNT (Partner): https://lmnt.com/\n\n\nBy Arxflix\n","html":"

DeepSeekMath: Revolutionizing Mathematical Reasoning in Open-Source AI

\n

\n\n

Links 🔗:

\n

👉 Subscribe: https://www.youtube.com/@Arxflix
👉 Twitter: https://x.com/arxflix
👉 LMNT (Partner): https://lmnt.com/

\n

By Arxflix

\n","updatedAt":"2024-06-09T01:46:22.680Z","author":{"_id":"6186ddf6a7717cb375090c01","avatarUrl":"/avatars/716b6a7d1094c8036b2a8a7b9063e8aa.svg","fullname":"Julien BLANCHON","name":"blanchon","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":142}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.5055437684059143},"editors":["blanchon"],"editorAvatarUrls":["/avatars/716b6a7d1094c8036b2a8a7b9063e8aa.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2402.03300","authors":[{"_id":"65c19a1f55c4f06fa9692c52","user":{"_id":"65db64f8b62d242ed8711701","avatarUrl":"/avatars/753e9f980eb6786c6b53b2f1becbf745.svg","isPro":false,"fullname":"Zhihong Shao","user":"ZhihongShao","type":"user"},"name":"Zhihong Shao","status":"claimed_verified","statusLastChangedAt":"2025-01-26T11:41:46.616Z","hidden":false},{"_id":"65c19a1f55c4f06fa9692c53","user":{"_id":"656873f33fd0bf1f82558695","avatarUrl":"/avatars/7a085da2e2a91d7f41988501a573ebf9.svg","isPro":false,"fullname":"PEIYI, WANG","user":"peiyiwang89","type":"user"},"name":"Peiyi Wang","status":"admin_assigned","statusLastChangedAt":"2024-02-13T08:13:15.437Z","hidden":false},{"_id":"65c19a1f55c4f06fa9692c54","user":{"_id":"63cd76b4374057a338e8e703","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63cd76b4374057a338e8e703/i4Qk5-0aYx3oRhC8b50aJ.jpeg","isPro":false,"fullname":"zhuqihao","user":"zqh11","type":"user"},"name":"Qihao Zhu","status":"admin_assigned","statusLastChangedAt":"2024-02-13T08:13:25.841Z","hidden":false},{"_id":"65c19a1f55c4f06fa9692c55","name":"Runxin Xu","hidden":false},{"_id":"65c19a1f55c4f06fa9692c56","user":{"_id":"6565a2dd131d13ccc5d8cb12","avatarUrl":"/avatars/f5c5441ba74791b64c9740911f952bac.svg","isPro":false,"fullname":"Junxiao Song","user":"haha-point","type":"user"},"name":"Junxiao Song","status":"admin_assigned","statusLastChangedAt":"2024-02-13T08:13:39.870Z","hidden":false},{"_id":"65c19a1f55c4f06fa9692c57","name":"Mingchuan Zhang","hidden":false},{"_id":"65c19a1f55c4f06fa9692c58","name":"Y. K. Li","hidden":false},{"_id":"65c19a1f55c4f06fa9692c59","name":"Y. Wu","hidden":false},{"_id":"65c19a1f55c4f06fa9692c5a","user":{"_id":"653df20eaa1f487614da4db1","avatarUrl":"/avatars/12b27ce2c59f53b7e464039deab36a5d.svg","isPro":false,"fullname":"Daya Guo","user":"guoday","type":"user"},"name":"Daya Guo","status":"admin_assigned","statusLastChangedAt":"2024-02-13T08:14:02.332Z","hidden":false}],"publishedAt":"2024-02-05T18:55:32.000Z","submittedOnDailyAt":"2024-02-06T00:57:34.722Z","title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open\n Language Models","submittedOnDailyBy":{"_id":"60f1abe7544c2adfd699860c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674929746905-60f1abe7544c2adfd699860c.jpeg","isPro":false,"fullname":"AK","user":"akhaliq","type":"user"},"summary":"Mathematical reasoning poses a significant challenge for language models due\nto its complex and structured nature. In this paper, we introduce DeepSeekMath\n7B, which continues pre-training DeepSeek-Coder-Base-v1.5 7B with 120B\nmath-related tokens sourced from Common Crawl, together with natural language\nand code data. DeepSeekMath 7B has achieved an impressive score of 51.7% on the\ncompetition-level MATH benchmark without relying on external toolkits and\nvoting techniques, approaching the performance level of Gemini-Ultra and GPT-4.\nSelf-consistency over 64 samples from DeepSeekMath 7B achieves 60.9% on MATH.\nThe mathematical reasoning capability of DeepSeekMath is attributed to two key\nfactors: First, we harness the significant potential of publicly available web\ndata through a meticulously engineered data selection pipeline. Second, we\nintroduce Group Relative Policy Optimization (GRPO), a variant of Proximal\nPolicy Optimization (PPO), that enhances mathematical reasoning abilities while\nconcurrently optimizing the memory usage of PPO.","upvotes":129,"discussionId":"65c19a2055c4f06fa9692c9b","ai_summary":"DeepSeekMath 7B improves mathematical reasoning through enhanced data pre-training and Group Relative Policy Optimization, achieving high scores on MATH benchmark without external tools.","ai_keywords":["DeepSeekMath 7B","DeepSeek-Coder-Base-v1.5 7B","MATH benchmark","Group Relative Policy Optimization","Proximal Policy Optimization","mathematical reasoning","data selection pipeline"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64b26c035e1230a79f897880","avatarUrl":"/avatars/5427b05b3ef627d4d8281f9a33bb98ab.svg","isPro":false,"fullname":"zhangwenbin","user":"ExceedZhang","type":"user"},{"_id":"6447384627a3e9e0b7c7b9aa","avatarUrl":"/avatars/5d0ae0d669554e5133c7b4f5ca83efb2.svg","isPro":false,"fullname":"Mr Jack Tung","user":"MrJackTung","type":"user"},{"_id":"647c84d9e07cf9bb2d467f69","avatarUrl":"/avatars/c66f2a4e0da3ba24aa7d4c050026fe6d.svg","isPro":false,"fullname":"Maksym Sutkovenko","user":"Subuday","type":"user"},{"_id":"60d418c8d7e9cf17e5265ae4","avatarUrl":"/avatars/9f1d9a44b79e0c81d594b50844f0aeda.svg","isPro":false,"fullname":"Qin","user":"Lonnie","type":"user"},{"_id":"6555125a4f361968f0e3aad7","avatarUrl":"/avatars/e7692d82804338f21ecdc6e731f5c5ea.svg","isPro":false,"fullname":"marinaretikof","user":"marinaretik","type":"user"},{"_id":"625026b7d2d191ac43320c5e","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/625026b7d2d191ac43320c5e/2ExzHlZ-Bk8SQMyBjeY6N.jpeg","isPro":false,"fullname":"Jingcheng Hu","user":"reign12","type":"user"},{"_id":"5e32d89653d2a72512789cdc","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5e32d89653d2a72512789cdc/anNXBAYRGuh9jFw2eOUEb.jpeg","isPro":false,"fullname":"Arunkumar Venkataramanan","user":"ArunkumarVR","type":"user"},{"_id":"654448bb26612e5469a4d672","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/1O3niyiqaKPTNkGpPJSKU.png","isPro":false,"fullname":"Marat Babaev","user":"maratFC","type":"user"},{"_id":"636cee7acfb49b46822006a1","avatarUrl":"/avatars/928b1cc75ff8e768bfe0cbdd40d7b11f.svg","isPro":false,"fullname":"Attila Lukács","user":"attilalukacs","type":"user"},{"_id":"6032802e1f993496bc14d9e3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","isPro":false,"fullname":"Omar Sanseviero","user":"osanseviero","type":"user"},{"_id":"632289a7909ac44b572bf51a","avatarUrl":"/avatars/7ebca310419ea7c3e0f5a4b57fbbdf6e.svg","isPro":false,"fullname":"galen","user":"AlphaGalen","type":"user"},{"_id":"64ca7c04710645aa7bdbbfff","avatarUrl":"/avatars/c12f4cb6dc1ff0010edb3ef4cfcccd7c.svg","isPro":false,"fullname":"Lize Pirenne","user":"Inversta","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":1}">