Lattice AI Research

Research focus

Reasoning & Chain-of-Thought (1)RLHF & Preference Learning (1)Data Curation & Synthetic Data (1)Recommendation & Information Retrieval (1)

Frequent co-authors

Mohammad Beigi (1)Lifu Huang (1)Wentao Hu (1)Zhendong Chu (1)

Papers (2)

Jun 8, 2026

1w ago·also Virginia Tech

Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization

PRIME reveals a crucial precursor to reward hacking that can predict and adapt to misalignment before it manifests, offering a new lens on alignment risks in RL systems.

Mohammad Beigi, Ming Jin, Lifu Huang

Reasoning & Chain-of-Thought RLHF & Preference Learning

May 28, 2026

Wentao Hu +83w ago·also SJTU, Virginia Tech

SkillBrew: Multi-Objective Curation of Skill Banks for LLM Agents

LLM agents can be significantly improved by *removing* redundant and outdated skills from their skill banks, not just adding more.

Wentao Hu, Zhendong Chu, Yiming Zhang +6

Data Curation & Synthetic Data Recommendation & Information Retrieval Tool Use & Agents

Search

Ming Jin

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)