Zidong Taichu Introduces GMC Core-Set Pruning: 80% Fewer Tokens, Full Multimodal Fidelity

The Zidong Taichu team at the Chinese Academy of Sciences has proposed GMC, a core-set pruning method that reduces visual tokens by over 80% while preserving multimodal performance, addressing the bottleneck of token redundancy. The training-free method works with mainstream models like Qwen and LLaVA, matching or exceeding original performance on multiple benchmarks.

+4