5 papers
TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agents
Jiaqi Wang, Wenhao Zhang, Weijie Shi +2
On-policy distillation (OPD) has shown strong potential for transferring reasoning ability from frontier or domain-specific models to smaller students. While effective on static si…
VideoASMR-Bench: Can AI-Generated ASMR Videos Fool VLMs and Humans?
Jiaqi Wang, Weijia Wu, Yi Zhan +6
With AI-generated videos increasingly indistinguishable from reality, current benchmarks primarily focus on broad semantic alignment and basic physical consistency, offering limite…
Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
Jiaqi Wang, Kevin Qinghong Lin, James Cheng +1
Reinforcement Learning (RL) has proven to be an effective post-training strategy for enhancing reasoning in vision-language models (VLMs). Group Relative Policy Optimization (GRPO)…
DivIL: Unveiling and Addressing Over-Invariance for Out-of- Distribution Generalization
Jiaqi Wang, Yuhang Zhou, Zhixiong Zhang +3
Out-of-distribution generalization is a common problem that expects the model to perform well in the different distributions even far from the train data. A popular approach to add…
A Signed Graph Approach to Understanding and Mitigating Oversmoothing in GNNs
Jiaqi Wang, Xinyi Wu, James Cheng +1
Deep graph neural networks (GNNs) often suffer from oversmoothing, where node representations become overly homogeneous with increasing depth. While techniques like normalization,…