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

The following papers were recommended by the Semantic Scholar API

Data-driven active learning approaches for accelerating materials discovery (2026)
AI Meets Plasticity: A Comprehensive Survey (2026)
The Path Ahead for Agentic AI: Challenges and Opportunities (2026)
MADE: Benchmark Environments for Closed-Loop Materials Discovery (2026)

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":"2026-02-11T01:44:39.280Z","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":0.7785104513168335},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.00169","authors":[{"_id":"698b5df76052d3bed9630816","name":"Huan Zhang","hidden":false},{"_id":"698b5df76052d3bed9630817","name":"Yizhan Li","hidden":false},{"_id":"698b5df76052d3bed9630818","name":"Wenhao Huang","hidden":false},{"_id":"698b5df76052d3bed9630819","name":"Ziyu Hou","hidden":false},{"_id":"698b5df76052d3bed963081a","name":"Yu Song","hidden":false},{"_id":"698b5df76052d3bed963081b","user":{"_id":"60845b6ca5da133ac6c38681","avatarUrl":"/avatars/01dfcf615a57c37ff19276d79f423cf1.svg","isPro":false,"fullname":"Xuye Liu","user":"xuyeliu123","type":"user"},"name":"Xuye Liu","status":"claimed_verified","statusLastChangedAt":"2026-02-11T11:15:56.294Z","hidden":false},{"_id":"698b5df76052d3bed963081c","name":"Farshid Effaty","hidden":false},{"_id":"698b5df76052d3bed963081d","user":{"_id":"645605624cfac492278dcd9c","avatarUrl":"/avatars/0c8c19fa3c58f05288bea47757548db5.svg","isPro":false,"fullname":"Yaya Jiang","user":"j9jiang","type":"user"},"name":"Jinya Jiang","status":"claimed_verified","statusLastChangedAt":"2026-02-11T11:15:58.893Z","hidden":false},{"_id":"698b5df76052d3bed963081e","name":"Sifan Wu","hidden":false},{"_id":"698b5df76052d3bed963081f","name":"Qianggang Ding","hidden":false},{"_id":"698b5df76052d3bed9630820","name":"Izumi Takahara","hidden":false},{"_id":"698b5df76052d3bed9630821","name":"Leonard R. MacGillivray","hidden":false},{"_id":"698b5df76052d3bed9630822","name":"Teruyasu Mizoguchi","hidden":false},{"_id":"698b5df76052d3bed9630823","name":"Tianshu Yu","hidden":false},{"_id":"698b5df76052d3bed9630824","name":"Lizi Liao","hidden":false},{"_id":"698b5df76052d3bed9630825","name":"Yuyu Luo","hidden":false},{"_id":"698b5df76052d3bed9630826","name":"Yu Rong","hidden":false},{"_id":"698b5df76052d3bed9630827","name":"Jia Li","hidden":false},{"_id":"698b5df76052d3bed9630828","name":"Ying Diao","hidden":false},{"_id":"698b5df76052d3bed9630829","name":"Heng Ji","hidden":false},{"_id":"698b5df76052d3bed963082a","name":"Bang Liu","hidden":false}],"publishedAt":"2026-01-29T23:48:43.000Z","submittedOnDailyAt":"2026-02-10T14:10:13.591Z","title":"Towards Agentic Intelligence for Materials Science","submittedOnDailyBy":{"_id":"658f310a89145cbc7c967e35","avatarUrl":"/avatars/9e3baa66f7292bb6d42670bc3246f6b7.svg","isPro":false,"fullname":"Huan Zhang","user":"Huan0825","type":"user"},"summary":"The convergence of artificial intelligence and materials science presents a transformative opportunity, but achieving true acceleration in discovery requires moving beyond task-isolated, fine-tuned models toward agentic systems that plan, act, and learn across the full discovery loop. This survey advances a unique pipeline-centric view that spans from corpus curation and pretraining, through domain adaptation and instruction tuning, to goal-conditioned agents interfacing with simulation and experimental platforms. Unlike prior reviews, we treat the entire process as an end-to-end system to be optimized for tangible discovery outcomes rather than proxy benchmarks. This perspective allows us to trace how upstream design choices-such as data curation and training objectives-can be aligned with downstream experimental success through effective credit assignment.\n To bridge communities and establish a shared frame of reference, we first present an integrated lens that aligns terminology, evaluation, and workflow stages across AI and materials science. We then analyze the field through two focused lenses: From the AI perspective, the survey details LLM strengths in pattern recognition, predictive analytics, and natural language processing for literature mining, materials characterization, and property prediction; from the materials science perspective, it highlights applications in materials design, process optimization, and the acceleration of computational workflows via integration with external tools (e.g., DFT, robotic labs). Finally, we contrast passive, reactive approaches with agentic design, cataloging current contributions while motivating systems that pursue long-horizon goals with autonomy, memory, and tool use. This survey charts a practical roadmap towards autonomous, safety-aware LLM agents aimed at discovering novel and useful materials.","upvotes":46,"discussionId":"698b5df86052d3bed963082b","ai_summary":"AI-driven materials science integrates large language models across discovery pipelines from data curation to agent-based experimentation, emphasizing system-level optimization and autonomous goal pursuit.","ai_keywords":["artificial intelligence","materials science","agentic systems","pipeline-centric view","domain adaptation","instruction tuning","goal-conditioned agents","simulation platforms","experimental platforms","credit assignment","LLMs","pattern recognition","predictive analytics","natural language processing","literature mining","materials characterization","property prediction","materials design","process optimization","computational workflows","DFT","robotic labs","passive approaches","reactive approaches","autonomous agents","safety-aware agents"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"645605624cfac492278dcd9c","avatarUrl":"/avatars/0c8c19fa3c58f05288bea47757548db5.svg","isPro":false,"fullname":"Yaya Jiang","user":"j9jiang","type":"user"},{"_id":"641e5bf65f274a0a92c2f6a2","avatarUrl":"/avatars/9c25593ed07dd0240870670d94c082ab.svg","isPro":false,"fullname":"Wenhao Huang","user":"EZ-hwh","type":"user"},{"_id":"658f310a89145cbc7c967e35","avatarUrl":"/avatars/9e3baa66f7292bb6d42670bc3246f6b7.svg","isPro":false,"fullname":"Huan Zhang","user":"Huan0825","type":"user"},{"_id":"654a97282d2fcd6bf2851173","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/654a97282d2fcd6bf2851173/9zXf940gr4WNt4e-oOt4k.png","isPro":false,"fullname":"Bang Liu","user":"Bang-UdeM-Mila","type":"user"},{"_id":"6506e375dacc94cd6c0b5193","avatarUrl":"/avatars/9966b872737a7172ea52af3206cdb0a7.svg","isPro":false,"fullname":"Jiaming Xu","user":"soxziw","type":"user"},{"_id":"63ccd72d336d9d56171634d9","avatarUrl":"/avatars/2cce0f1eeaaf25f1382face573db5921.svg","isPro":false,"fullname":"Lingqi Z","user":"Lynxintherye","type":"user"},{"_id":"620d5b00d6db8f314af2dd5f","avatarUrl":"/avatars/cc0329170fa1ef68e98893de2066c92d.svg","isPro":false,"fullname":"HAOCHEN SHI","user":"hcshi","type":"user"},{"_id":"648c203f2b9b7c6a99181dde","avatarUrl":"/avatars/f1e95a8b40693de1529a5f0b2190901d.svg","isPro":false,"fullname":"Xiran Song","user":"XiranSong","type":"user"},{"_id":"655c092183186f133f959108","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/WMVgGhjQbVJXK9eh4EuT9.jpeg","isPro":false,"fullname":"Xiaoqiang Wang","user":"qindomitable","type":"user"},{"_id":"64c090a9f613170e7be93d2f","avatarUrl":"/avatars/ccbdf444e1f2386d2281e8e42059ebb0.svg","isPro":false,"fullname":"KunlunZhu","user":"KunlunZhu","type":"user"},{"_id":"66326a6ba6ef9d3afd025fd2","avatarUrl":"/avatars/456eb1ca0a702fcb50031f6539ae78ab.svg","isPro":false,"fullname":"Hou","user":"ZiyuPeter","type":"user"},{"_id":"6552817e9e144c06dd0d66f1","avatarUrl":"/avatars/d93552c0ef9ae697ec416ef4191957cb.svg","isPro":false,"fullname":"Qian Yang","user":"QianYangMILA","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2602.00169

Towards Agentic Intelligence for Materials Science

Published on Jan 29

· Submitted by

Huan Zhang on Feb 10

Upvote

Authors:

Xuye Liu ,

Jinya Jiang ,

Abstract

AI-driven materials science integrates large language models across discovery pipelines from data curation to agent-based experimentation, emphasizing system-level optimization and autonomous goal pursuit.

AI-generated summary

The convergence of artificial intelligence and materials science presents a transformative opportunity, but achieving true acceleration in discovery requires moving beyond task-isolated, fine-tuned models toward agentic systems that plan, act, and learn across the full discovery loop. This survey advances a unique pipeline-centric view that spans from corpus curation and pretraining, through domain adaptation and instruction tuning, to goal-conditioned agents interfacing with simulation and experimental platforms. Unlike prior reviews, we treat the entire process as an end-to-end system to be optimized for tangible discovery outcomes rather than proxy benchmarks. This perspective allows us to trace how upstream design choices-such as data curation and training objectives-can be aligned with downstream experimental success through effective credit assignment. To bridge communities and establish a shared frame of reference, we first present an integrated lens that aligns terminology, evaluation, and workflow stages across AI and materials science. We then analyze the field through two focused lenses: From the AI perspective, the survey details LLM strengths in pattern recognition, predictive analytics, and natural language processing for literature mining, materials characterization, and property prediction; from the materials science perspective, it highlights applications in materials design, process optimization, and the acceleration of computational workflows via integration with external tools (e.g., DFT, robotic labs). Finally, we contrast passive, reactive approaches with agentic design, cataloging current contributions while motivating systems that pursue long-horizon goals with autonomy, memory, and tool use. This survey charts a practical roadmap towards autonomous, safety-aware LLM agents aimed at discovering novel and useful materials.