Guiguang Ding

Unbiased Token Pruning for Efficient Large Multimodal Models

mmSSR: Harvesting Rich, Scalable and Transferable Multi-Modal Data for Instruction Fine-Tuning

PruneHal: Reducing Hallucinations in Multi-modal Large Language Models through Adaptive KV Cache Pruning

FastVID: Dynamic Density Pruning for Fast Video Large Language Models

PrefixKV: Adaptive Prefix KV Cache is What Vision Instruction-Following Models Need for Efficient Generation

TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval

Towards Efficient Vision-Language Tuning: More Information Density, More Generalizability