Lehigh AI Research Lab
We are LAIR at Lehigh University, led by Prof. Lichao Sun. We study how to make AI systems more reliable, robust, efficient, and interpretable, and build tools for AI-assisted scientific research.
Lab website · GitHub · Publications · People
Research areas
Trustworthy AI · Medical AI · Generative AI and large language models · Federated learning · Reinforcement learning
Our recent projects explore scientific agents, looped decision models, and the evaluation of human and AI-generated peer reviews.
Models and datasets
The following resources accompany our research projects.
| Project | Resources | What you will find |
|---|---|---|
| LEV | Lev-1.4B · Lev-2.6B · Lev-3B | Looped decision models with multiple inference depths. |
| LEV evaluation | LevBench | Development and locked-test decisions across five tiers; a rights-filtered subset. |
| LEV training | Lev-train-v2 | Training mixtures and recipes; a rights-filtered subset. |
| LEV results | Lev-results | Per-depth outputs, calibration, and latency evidence, without question text. |
| PremiseCheck | ReviewFact | A benchmark for factual errors in peer-review comments. For evaluation, not model training. |
Explore the LEV collection, LEV code, and PremiseCheck code. Please consult each resource card for its intended use, limitations, and license terms.
Research tools and projects
- Dr. Claw — An AI research workspace for organizing agent-assisted scientific workflows.
- OpenSkill — Research on agents that develop skills and verification signals through self-evolution.
- ClawChain — Runtime monitoring, recovery, and evidence tracking for AI agents.
- Awesome Vibe Researching — A curated collection of resources for AI-driven research automation.
For research updates, collaboration, and open positions, visit our lab website.