4 papers
Reinforcement Learning via Value Gradient Flow
Haoran Xu, Kaiwen Hu, Somayeh Sojoudi +1
We study behavior-regularized reinforcement learning (RL), where regularization toward a reference distribution (the dataset in offline RL or the base model in LLM RL finetuning) i…
MM-Doc-R1: Training Agents for Long Document Visual Question Answering through Multi-turn Reinforcement Learning
Jiahang Lin, Kai Hu, Binghai Wang +12
Conventional Retrieval-Augmented Generation (RAG) systems often struggle with complex multi-hop queries over long documents due to their single-pass retrieval. We introduce MM-Doc-…
Projection Head is Secretly an Information Bottleneck
Zhuo Ouyang, Kaiwen Hu, Qi Zhang +2
Recently, contrastive learning has risen to be a promising paradigm for extracting meaningful data representations. Among various special designs, adding a projection head on top o…
Understanding the Role of Equivariance in Self-supervised Learning
Yifei Wang, Kaiwen Hu, Sharut Gupta +3
Contrastive learning has been a leading paradigm for self-supervised learning, but it is widely observed that it comes at the price of sacrificing useful features (\eg colors) by b…