7 papers
Exploiting Verification-Generation Gap: Test-Time Reinforcement Learning with Confidence-Conditioned Verification
Jiahui Li, Jianfeng Shan, Wenpei Chen +5
Test-time reinforcement learning has emerged as a promising paradigm for enhancing the complex reasoning abilities of large language models in a completely label-free manner. Despi…
Path-Decoupled Hyperbolic Flow Matching for Few-Shot Adaptation
Lin Li, Ziqi Jiang, Gefan Ye +5
Recent advances in cross-modal few-shot adaptation treat visual-semantic alignment as a continuous feature transport problem via Flow Matching (FM). However, we argue that Euclidea…
em: Learning Hierarchical Hyperbolic Embeddings for Compositional Zero-Shot Learning
Lin Li, Jiahui Li, Jiaming Lei +3
Compositional zero-shot learning (CZSL) aims to recognize unseen state-object compositions by generalizing from a training set of their primitives (state and object). Current metho…
Relation-R1: Progressively Cognitive Chain-of-Thought Guided Reinforcement Learning for Unified Relation Comprehension
Lin Li, Wei Chen, Jiahui Li +2
Recent advances in multi-modal large language models (MLLMs) have significantly improved object-level grounding and region captioning. However, they remain limited in visual relati…
RED: Unleashing Token-Level Rewards from Holistic Feedback via Reward Redistribution
Jiahui Li, Lin Li, Tai-wei Chang +4
Reinforcement learning from human feedback (RLHF) offers a promising approach to aligning large language models (LLMs) with human preferences. Typically, a reward model is trained…
Towards Better Alignment: Training Diffusion Models with Reinforcement Learning Against Sparse Rewards
Zijing Hu, Fengda Zhang, Long Chen +6
Diffusion models have achieved remarkable success in text-to-image generation. However, their practical applications are hindered by the misalignment between generated images and c…