一作NeurIPS 2026Poster
Beyond Prediction: Steering VLM Agents with Retrospective World Modeling
Yongjiang Liu · Jie ZHANG · Haoyue Zhang · Jingcai Guo · Deze Zeng · Song Guo
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Yongjiang Liu · Jie ZHANG · Haoyue Zhang · Jingcai Guo · Deze Zeng · Song Guo
Yongjiang Liu · Haoxi Li · Xiaosong Ma · Jie ZHANG · Song Guo
Recent Large Reasoning Models (LRMs) excel at complex reasoning tasks but often suffer from overthinking, generating overly long and redundant reasoning trajectories. To explore its essence, our empirical analysis reveals that LRMs are primarily limited to recognizing task properties (i.e., diffi...
Sikai Bai · Haoxi Li · Jie ZHANG · Yongjiang Liu · Song Guo
Despite significant advances in Large Reasoning Models (LRMs) driven by reinforcement learning with verifiable rewards (RLVR), this paradigm is fundamentally limited in specialized or novel domains where such supervision is prohibitively expensive or unavailable, posing a key challenge for test-t...