Submitted by Yif Yang 11 BizGenEval: A Systematic Benchmark for Commercial Visual Content Generation Microsoft Research 22 2
Submitted by JeonghyeKim 58 Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs? Microsoft Research 75 7
Submitted by JeonghyeKim 12 Understanding Reasoning in LLMs through Strategic Information Allocation under Uncertainty Microsoft Research 8 2
Submitted by Zongqian Li 5 Scaling Data Difficulty: Improving Coding Models via Reinforcement Learning on Fresh and Challenging Problems Microsoft Research 2
Submitted by Zongqian Li 5 Breaking Training Bottlenecks: Effective and Stable Reinforcement Learning for Coding Models Microsoft Research 11 2
Submitted by Akshay Nambi 17 Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces Microsoft Research 3
Submitted by ZD 6 Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity Microsoft Research 17 2
Submitted by taesiri 37 Proact-VL: A Proactive VideoLLM for Real-Time AI Companions Microsoft Research 137 8
Submitted by Akshay Nambi 13 Learning When to Act or Refuse: Guarding Agentic Reasoning Models for Safe Multi-Step Tool Use Microsoft Research 3
Submitted by Baolin Peng 28 Reinforcement World Model Learning for LLM-based Agents Microsoft Research 4
Submitted by junchao-cuhk 14 LIVE: Long-horizon Interactive Video World Modeling Microsoft Research 41 3