Seanie Lee

Rethinking Reward Models for Multi-Domain Test-Time Scaling

HoliSafe: Holistic Safety Benchmarking and Modeling for Vision-Language Model

FedSVD: Adaptive Orthogonalization for Private Federated Learning with LoRA

Reliable Decision‑Making via Calibration‑Oriented Retrieval‑Augmented Generation

Distilling LLM Agent into Small Models with Retrieval and Code Tools

Trajectory Balance with Asynchrony: Decoupling Exploration and Learning for Fast, Scalable LLM Post-Training

Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning

HarmAug: Effective Data Augmentation for Knowledge Distillation of Safety Guard Models

Calibrated Decision-Making through Large Language Model-Assisted Retrieval