PPL
Jian Yang
Papers (9)
Qwen2.5-Coder Technical ReportMIRA: Mid-training Rubric Anchoring for Source-Aware Data SelectionRethinking Stepwise Model Routing: A Cost-Efficient Table Reasoning PerspectiveRubric-Guided Process Reward for Stepwise Model RoutingRW-TTT: Batched Serving for Request-Owned Test-Time Training StateHTMLCure: Turning Browser Experience into State Guided Repair for Interactive HTMLLANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint GuidanceDistill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous DrivingRobust LLM Unlearning Against Relearning Attacks: The Minor Components in Representations Matter