Songyang Gao

Venues

Frequent coauthors

Papers

1. AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse Environments
   ACL 2025 ·

   Zhiheng Xi

   DBLP profile ↗ORCID search ↗
2. Alleviating Shifted Distribution in Human Preference Alignment through Meta-Learning
   AAAI 2025 ·

   Shihan Dou

   DBLP profile ↗ORCID search ↗
3. Are Your LLMs Capable of Stable Reasoning?
   ACL 2025 ·

   Junnan Liu

   DBLP profile ↗ORCID search ↗
4. Capability Salience Vector: Fine-grained Alignment of Loss and Capabilities for Downstream Task Scaling Law
   ACL 2025 ·

   Qiming Ge

   DBLP profile ↗ORCID search ↗
5. CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome Reward
   EMNLP 2025 ·

   Shudong Liu

   DBLP profile ↗ORCID search ↗
6. Pre-Trained Policy Discriminators are General Reward Models
   NeurIPS 2025 ·

   Shihan Dou

   DBLP profile ↗ORCID search ↗
7. Semi-off-Policy Reinforcement Learning for Vision-Language Slow-Thinking Reasoning
   NeurIPS 2025 ·

   Junhao Shen

   DBLP profile ↗ORCID search ↗
8. Inverse-Q*: Token Level Reinforcement Learning for Aligning Large Language Models Without Preference Data
   EMNLP 2024 ·

   Han Xia

   DBLP profile ↗ORCID search ↗
9. Linear Alignment: A Closed-form Solution for Aligning Human Preferences without Tuning and Feedback
   ICML 2024 · Songyang Gao
10. LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
    ACL 2024 ·

    Shihan Dou

    DBLP profile ↗ORCID search ↗
11. Navigating the OverKill in Large Language Models
    ACL 2024 ·

    Chenyu Shi

    DBLP profile ↗ORCID search ↗
12. RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning
    EMNLP 2024 ·

    Junjie Ye

    DBLP profile ↗ORCID search ↗
13. ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages
    ACL 2024 ·

    Junjie Ye

    DBLP profile ↗ORCID search ↗