影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
46.01/100
前 10.4%
全站排名 #6,677
发表论文10 篇
平均评分
年均产出3.3 篇/年
Songyang Gao
研究方向
Reasoning · LLM Alignment · LLM Pretraining · Robustness
33
Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning
ICLR 2026Poster
22
OPV: Outcome-based Process Verifier for Efficient Long Chain-of-Thought Verification
ICLR 2026Rejected
10
LONG-HORIZON REASONING AGENT FOR OLYMPIAD- LEVEL MATHEMATICAL PROBLEM SOLVING
ICLR 2026Rejected
一作13
The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner
ICLR 2026Poster
24
Pre-Trained Policy Discriminators are General Reward Models
NeurIPS 2025Poster
29
Semi-off-Policy Reinforcement Learning for Vision-Language Slow-Thinking Reasoning
NeurIPS 2025Poster
16
Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning
COLM 2025Poster
二作44
AgentGym: Evaluating and Evolving Large Language Model-based Agents across Diverse Envronments
ICLR 2025Rejected