PPL
Bo Han
Papers (10)
AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment CorruptionsHidden Human-Like Nature of Machine-Generated Texts: Theory and Detection EnhancementConditional Equivalence of DPO and RLHF: Implicit Assumption, Failure Modes, and Provable AlignmentWhat Do Evolutionary Coding Agents Evolve?Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent MemoryDistinguishable Deletion: Unifying Knowledge Erasure and Refusal for Large Language Model UnlearningMulti-Level Contextual Token Relation Modeling for Machine-Generated Text DetectionShaping Schema via Language Representation as the Next Frontier for LLM Intelligence ExpandingBeyond Distribution Estimation: Simplex Anchored Structural Inference Towards Universal Semi-Supervised LearningBeyond Rigid Alignment: Graph Federated Learning via Dual Manifold Calibration