Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456 Paper page - CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion
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":"2026-02-13T01:43:11.659Z","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":0.7210378050804138},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.10999","authors":[{"_id":"698d437e65c0d15a6d162132","name":"Yusong Lin","hidden":false},{"_id":"698d437e65c0d15a6d162133","user":{"_id":"65f43c3cc9940817caaf4434","avatarUrl":"/avatars/ecec2856ba7a7d3421a2071a0a88800b.svg","isPro":false,"fullname":"Haiyang Wang","user":"Haiyang-W","type":"user"},"name":"Haiyang Wang","status":"claimed_verified","statusLastChangedAt":"2026-02-12T13:28:33.349Z","hidden":false},{"_id":"698d437e65c0d15a6d162134","name":"Shuzhe Wu","hidden":false},{"_id":"698d437e65c0d15a6d162135","name":"Lue Fan","hidden":false},{"_id":"698d437e65c0d15a6d162136","name":"Feiyang Pan","hidden":false},{"_id":"698d437e65c0d15a6d162137","name":"Sanyuan Zhao","hidden":false},{"_id":"698d437e65c0d15a6d162138","name":"Dandan Tu","hidden":false}],"publishedAt":"2026-02-11T16:22:18.000Z","submittedOnDailyAt":"2026-02-12T00:35:37.899Z","title":"CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},"summary":"Agentic coding requires agents to effectively interact with runtime environments, e.g., command line interfaces (CLI), so as to complete tasks like resolving dependency issues, fixing system problems, etc. But it remains underexplored how such environment-intensive tasks can be obtained at scale to enhance agents' capabilities. To address this, based on an analogy between the Dockerfile and the agentic task, we propose to employ agents to simulate and explore environment histories, guided by execution feedback. By tracing histories of a healthy environment, its state can be inverted to an earlier one with runtime failures, from which a task can be derived by packing the buggy state and the corresponding error messages. With our method, named CLI-Gym, a total of 1,655 environment-intensive tasks are derived, being the largest collection of its kind. Moreover, with curated successful trajectories, our fine-tuned model, named LiberCoder, achieves substantial absolute improvements of +21.1% (to 46.1%) on Terminal-Bench, outperforming various strong baselines. To our knowledge, this is the first public pipeline for scalable derivation of environment-intensive tasks.","upvotes":10,"discussionId":"698d437e65c0d15a6d162139","githubRepo":"https://github.com/LiberCoders/CLI-Gym","githubRepoAddedBy":"user","ai_summary":"CLI-Gym enables scalable derivation of environment-intensive tasks by simulating and exploring environment histories, while LiberCoder achieves significant performance improvements on Terminal-Bench through fine-tuning.","ai_keywords":["CLI-Gym","LiberCoder","environment-intensive tasks","Dockerfile","agentic task","execution feedback","Terminal-Bench"],"githubStars":10},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"646b43deb1202bc77c1024a4","avatarUrl":"/avatars/cf791574ab986bac274e7fbcf04e2a59.svg","isPro":false,"fullname":"hangyu guo","user":"Rosiness","type":"user"},{"_id":"6868f58a4757672a6da7c417","avatarUrl":"/avatars/73154b7e0f1af68054b97f10a6c2e670.svg","isPro":false,"fullname":"JiaCheng Zhang","user":"jiachengzhg","type":"user"},{"_id":"649ecf9827145c4463240177","avatarUrl":"/avatars/27696cf31790a3d58d8be2e0c983800e.svg","isPro":false,"fullname":"Lue Fan","user":"Abyssaledge","type":"user"},{"_id":"695f8d5d9e778cf056e56017","avatarUrl":"/avatars/a1ec00149303e4c69e43821d2ee43218.svg","isPro":false,"fullname":"Xiaoche","user":"x1aoche","type":"user"},{"_id":"639a8f29b2740bf1474e82c1","avatarUrl":"/avatars/306ac149819c80b66386e4a719662130.svg","isPro":false,"fullname":"Hongbo Wang","user":"Larer","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":"65f43c3cc9940817caaf4434","avatarUrl":"/avatars/ecec2856ba7a7d3421a2071a0a88800b.svg","isPro":false,"fullname":"Haiyang Wang","user":"Haiyang-W","type":"user"},{"_id":"666aa99cd1652853e4f9a8b9","avatarUrl":"/avatars/7cd5a0c34b5ccb8eff5a353d88d15a93.svg","isPro":false,"fullname":"HanXiao","user":"HanXiao1999","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":"698e5f796150030676e603d9","avatarUrl":"/avatars/a6e51a0fbe6a193e5277899827e54c75.svg","isPro":false,"fullname":"Anna-MariaHernandez","user":"goosemaria149","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">
CLI-Gym enables scalable derivation of environment-intensive tasks by simulating and exploring environment histories, while LiberCoder achieves significant performance improvements on Terminal-Bench through fine-tuning.
AI-generated summary
Agentic coding requires agents to effectively interact with runtime environments, e.g., command line interfaces (CLI), so as to complete tasks like resolving dependency issues, fixing system problems, etc. But it remains underexplored how such environment-intensive tasks can be obtained at scale to enhance agents' capabilities. To address this, based on an analogy between the Dockerfile and the agentic task, we propose to employ agents to simulate and explore environment histories, guided by execution feedback. By tracing histories of a healthy environment, its state can be inverted to an earlier one with runtime failures, from which a task can be derived by packing the buggy state and the corresponding error messages. With our method, named CLI-Gym, a total of 1,655 environment-intensive tasks are derived, being the largest collection of its kind. Moreover, with curated successful trajectories, our fine-tuned model, named LiberCoder, achieves substantial absolute improvements of +21.1% (to 46.1%) on Terminal-Bench, outperforming various strong baselines. To our knowledge, this is the first public pipeline for scalable derivation of environment-intensive tasks.