Progressive Misalignment Can Emerge from Benign Recursive Self-Distillation
In submission
Workshop on Transitioning from Pre-Training to Post-Training at NeurIPS 2026 (Oral)
Publications
2026
Learning the Preferences of a Learning Agent
Workshop on Multi-Agent Learning and Its Opportunities in the Era of Generative AI at ICLR 2026
2025
2024
Dynamic Vocabulary Pruning in Early-Exit LLMs
Workshop on Efficient Natural Language and Speech Processing at NeurIPS 2024
Algorithms for Caching and MTS with reduced number of predictions
ICLR 2024
'Explaining RL Decisions with Trajectories': A Reproducibility Study
TMLR 2024