Minjia Zhang

From Collapse to Control: Understanding and Extending Context Length in Emerging Hybrid Models via Universal Position Interpolation

VTool-R1: VLMs Learn to Think with Images via Reinforcement Learning on Multimodal Tool Use

AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism

Slow-Fast Policy Optimization: Reposition-Before-Update for LLM Reasoning

PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models via Sparse Expert Merging and Bit-packed inference

Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment

Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks