AI.info

Luke Zettlemoyer

Explore Luke Zettlemoyer on AI.info.

• Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall
• SPADE: Self-Play in Adaptive Synthetic Executable Environments
• WaiT for the Signal: Simple Frequency-Aware Flow-Matching
• Anchored Decoding: Provably Reducing Copyright Risk for Any Language Model
• MoCo: A One-Stop Shop for Model Collaboration Research
• Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image
• TV2TV: A Unified Framework for Interleaved Language and Video Generation
• DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research