AI.info

Zheng Li

Explore Zheng Li on AI.info.

• HeaPA: Difficulty-Aware Heap Sampling and On-Policy Query Augmentation for LLM Reinforcement Learning
• Compensating Distribution Drifts in Class-incremental Learning of Pre-trained Vision Transformers
• Self-Rewarding PPO: Aligning Large Language Models with Demonstrations Only
• Instant Personalized Large Language Model Adaptation via Hypernetwork
• DeepPlanner: Scaling Planning Capability for Deep Research Agents via Advantage Shaping