13 papers
DRDN: Decoupled Representation Dynamic Network for From-Scratch ViT Class-Incremental Learning
Bingchen Huang, Yifu Chen, Zhiling Wang +1
Dynamic expansion methods for class-incremental learning (CIL) protect task-specific knowledge by growing dedicated tokens or subnetworks, yet our analyses suggest that classificat…
Retrieved Images as Visual Thought: Training-Free Multimodal In-Context Learning for the Open-vs-Closed Gap
Bingchen Huang, Zhiling Wang, Yifu Chen +1
Recent work on Thinking with Images makes vision a dynamic part of reasoning, but does so through generation: the model invokes external tools, synthesizes code, or imagines new im…
On the Geometry of On-Policy Distillation
Zhennan Shen, Yanshu Li, Qingyu Yin +6
On-policy distillation (OPD) is increasingly used to improve large language model reasoning, but its training dynamics remain poorly understood. We characterize the trajectory of O…
Diffusion Forcing Planner: History-Annealed Planning with Time-Dependent Guidance for Autonomous Driving
Zehan Zhang, Neng Zhang, Yaoyi Li +2
Learning-based motion planners, despite recent progress, often suffer from temporal inconsistency. Small perturbations across frames can accumulate into unstable trajectories, degr…
Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
Hengyu Shi, Tianyang Han, Peizhe Wang +3
LLM post-training typically propagates task gradients through the full depth of the model. Although this end-to-end structure is simple and general, it couples task adaptation to f…
Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards
Tianyang Han, Hengyu Shi, Junjie Hu +3
Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal wh…