Cheng Deng

SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs

VoG: Enhancing LLM Reasoning through Stepwise Verification on Knowledge Graphs

Grouped-head latenT Attention

LoopServe: An Adaptive Dual-phase LLM Inference Acceleration System for Multi-Turn Dialogues

Activation-aware Probe-Query: Effective Key-Value Retrieval for Long-Context LLMs Inference

Pushing LLMs to Their Logical Reasoning Bound: The Role of Data Reasoning Intensity

MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework

NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens