AI.info

Pengfei Liu

Explore Pengfei Liu on AI.info.

• LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line Interfaces
• What Does Vision Tool-Use Reinforcement Learning Really Learn? Disentangling Tool-Induced and Intrinsic Effects for Crop-and-Zoom
• AgencyBench: Benchmarking the Frontiers of Autonomous Agents in 1M-Token Real-World Contexts
• SCALE: Selective Resource Allocation for Overcoming Performance Bottlenecks in Mathematical Test-time Scaling
• On Evaluating LLM Alignment by Evaluating LLMs as Judges
• Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresight
• SR-Scientist: Scientific Equation Discovery With Agentic AI