Talks

Credit Assignment: From Subgoal Confidence Signals to Recovery Harness for Long-Horizon Agents

Process Evaluation and Reliable Tool Use for Geometry Agents

Pipeline Design Principles of Agentic RAG

Emergence and Challenges of Vision-Language-Action (VLA) Systems: From PaLI-X and PaLM-E to RT-2 and cuRobo

Curated Data Selection in LLM Reasoning & Tree-structure Automated Agentic Workflow Design

Experience and Reflection on Recent LLM Reinforcement Learning with Verifiable Reward (RLVR)

Diffusion Model as Representation Learner (ICCV 2023)

Diversity-Driven Synthesis: Enhancing Dataset Distillation through Directed Weight Adjustment (Short Roadshow)

Towards Enhancing Diversity of Synthetic Data Distribution in ImageNet-1k-scale Dataset Distillation

Bi-Trajectory Guided Vision-Language Co-Distillation