Interactive Memory & Recursive Self-Improvement (RSI) for World Models
Project Lead · World model evaluation, scalable data curation, training, self-improvement. Ongoing.
MEIHAO FAN
Projects
EvoCost & SkillAdam: Agent Self-Evolution
EvoCost: Lead, First Author · SkillAdam: Co-author. ICLR 2027 submissions.
DeepAnalyze: Scalable Training-Trajectory Synthesis
Project Lead · Synthetic data, quality filtering, curriculum learning, agentic RL. ICML 2026.
CoDA-Bench & TACO: Scalable, Verifiable Task Construction
Co-author · Task synthesis, execution verification, difficulty refinement. ICML / VLDB 2026.
DeepPrep: Verifiable Task Synthesis for Agent Training
Project Lead; First Author · Executable supervision, reversible corruption, multi-turn RL. VLDB 2026.
AutoPrep: Multi-Agent Data Preparation Framework
A multi-agent data preparation framework with planner and programmer agents.
BATCHER: Cost-Effective LLM Inference for Entity Resolution
A cost-effective in-context learning framework for entity resolution.