Hao Peng

6 papers Department of Computer Science, University of Illinois Urbana-Champaign

• AGENTIF: Benchmarking Large Language Models Instruction Following Ability in Agentic Scenarios
• CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment
• Reinforcement Learning Finetunes Small Subnetworks in Large Language Models
• Structural Information-based Hierarchical Diffusion for Offline Reinforcement Learning
• The Best Instruction-Tuning Data are Those That Fit
• The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning