LLM Agents Can See Code Repositories
Paper ā¢ 2606.14061 ā¢ Published ā¢ 13
Lihan
SJTU-furrina
Ā·
AI & ML interests
None yet
Recent Activity
upvoted a paper 2 days ago
LLM Agents Can See Code Repositories upvoted a paper 8 days ago
SWE-Explore: Benchmarking How Coding Agents Explore Repositories upvoted a paper 5 months ago
SWE-Pruner: Self-Adaptive Context Pruning for Coding AgentsOrganizations
upvoted a paper 2 days ago
upvoted a paper 8 days ago
upvoted 3
papers 5 months ago
SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
Paper ā¢ 2601.16746 ā¢ Published ā¢ 92
Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning
Paper ā¢ 2601.06943 ā¢ Published ā¢ 215
GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts
Paper ā¢ 2601.05110 ā¢ Published ā¢ 29
upvoted a paper 6 months ago
reacted to YerbaPage's post with š„ 8 months ago
Post
2960
Latest work on SWE-Bench š
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp
upvoted a paper 9 months ago
reacted to YerbaPage's post with šš¤ 10 months ago
Post
2960
Latest work on SWE-Bench š
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp
reacted to YerbaPage's post with š 11 months ago
Post
2960
Latest work on SWE-Bench š
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp
Our two new papers from the SJTU & Huawei: Powered by DeepSeek-V3, we've achieved a new SOTA on the SWE-Bench benchmark!
We introduce two innovative approaches:
āļø SWE-Debate: AI agents compete and "debate" to generate the best code fix.
š§ SWE-Exp: An AI agent learns from past repair "experience" to solve new issues more efficiently.
š Explore the future of software development:
SWE-Debate
š Paper: https://arxiv.org/abs/2507.23348
š» Code: https://github.com/YerbaPage/SWE-Debate
SWE-Exp
š Paper: https://arxiv.org/abs/2507.23361
š» Code: https://github.com/YerbaPage/SWE-Exp