PPL
Jun Li
Papers (6)
Error as a Lens: Probing LLM Reasoning through Synthetic Misconception GenerationText-Preserving Lossy Text Compression: A Study of Strategic Deletion and LLM ReconstructionRealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior ImitationMulti-Scale Dequant: Eliminating Dequantization Bottleneck via Activation Decomposition for Efficient LLM InferenceU-STS-LLM A Unified Spatio-Temporal Steered Large Language Model for Traffic Prediction and ImputationRevisiting Uncertainty: On Evidential Learning for Partially Relevant Video Retrieval