Yu-Xin Xiao

Affiliation:

Papers on Lattice

Total citations

Topics

h-index

Publication activitypapers/week, last 8 weeks

Research focus

Scalable Oversight & Alignment Theory (1)Tool Use & Agents (1)Training Efficiency & Optimization (1)Eval Frameworks & Benchmarks (1)World Models & Planning (1)

Frequent co-authors

Zhi-Yun Lu (1)Jian-Yu Wang (1)Yang-Tian Zhang (1)Yang Song (1)Qi-Fan Wang (1)Bai-Lin Wang (1)Xiaomin Li (1)Jianheng Hou (1)

Papers (2)

Oct 8, 2026

Affiliation:6d ago

Recurrent Harness for Agentic Continual Learning

LLM agents can learn from past mistakes by retroactively revising their internal "harness" state, significantly improving continual learning performance.

Yu-Xin Xiao, Zhi-Yun Lu, Jian-Yu Wang +4

Scalable Oversight & Alignment Theory Tool Use & Agents Training Efficiency & Optimization

Aug 4, 2026

Affiliation:Aug 4, 2026·also MIT CSAIL, NVIDIA, Tsinghua AI, Great Bay University +8

MatrAIx: Simulating the World with 8.3 Billion Persona Agents

Simulating 8.3 billion diverse personas reveals nuanced user interactions that traditional evaluations miss, transforming how we assess AI systems.

Xiaomin Li, Jianheng Hou, Jintao Huang +82

Eval Frameworks & Benchmarks World Models & Planning

Yu-Xin Xiao

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)

Search

Yu-Xin Xiao

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)