PPL
Peng Wang
Papers (7)
Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and BeyondQwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any ResolutionRethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model CapabilityAgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and SecurityTowards Human-Like Interactive Speech Recognition With Agentic Correction and Semantic EvaluationFinBoardBench: Benchmarking Dynamic Wealth Management and Strategic Financial Reasoning of LLMs via Board Game SimulationsIs One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory