Wenhao Li

Out of the Memory Barrier: A Highly Memory-Efficient Training System for LLMs with Million-Token Contexts

SpaCE-10: A Comprehensive Benchmark for Multimodal Large Language Models in Compositional Spatial Intelligence

MetaCaptioner: Towards Generalist Visual Captioning with Open-source Suites

HS-SFT: Hybrid Sparse Supervised Fine-tuning for Offline LLM KV Cache Eviction

Beyond Proxy Metrics: A New Evaluation Framework for LLM Compression by Directly Measuring Generative Faithfulness

ChainGeo: Enabling Effective Geometric Reasoning in Small VLMs through Interleaved Visual-Text Chains

Spotlight Attention: Towards Efficient LLM Generation via Non-linear Hashing-based KV Cache Retrieval

Hierarchy-Aided Sparse Attention For Fast LLMs Prefilling Inference

SOLOS: Sparse Optimization For Long Sequence In Context Compression Enhanced LLMs