PPL
Weiqiang Wang
Papers (4)
Single-Rollout Hidden-State Dynamics for Training-Free RLVR Data SelectionEgoBench: An Interactive Egocentric Multimodal Benchmark for Tool-Using AgentsROAD: Adaptive Data Mixing for Offline-to-Online Reinforcement Learning via Bi-Level OptimizationTabEmbed: Benchmarking and Learning Generalist Embeddings for Tabular Understanding