PAR$^2$-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering
粗召回 & 细检索的多条 RAG 方案
Blogs
Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning
如何蒸馏图的推理路径
SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
SWE-Bench Pro
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
GRPO: Group Relative Policy Optimization
Proximal Policy Optimization Algorithms
PPO
REINFORCE Algorithm
著名的 REINFORCE 算法
Trust Region Policy Optimization
PPO 的前身
SWE-smith: Scaling Data for Software Engineering Agents
自动化 Repository-level 实例任务生成
SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
定制化 Agent-Computer Interface
SWE-Bench: Can Language Models Resolve Real-World GitHub Issues?
SWE-Bench