Tianfan Peng, Shijie Dong, Pengzhou Ji, Kailin Jiang, Minmin Ma, Yijun Tian, Jinhe Bi, Jinhe Bi, Wei Du, Feng Xiao, Lizhen Cui, Yuntao Du · ACM International Conference on Multimedia (ACM MM) 2026 · 2026
DOI: 10.1145/3767308.3835379
Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
Large multimodal models (LMMs) often suffer from severe inference inefficiency due to the large number of visual tokens introduced by image encoders. While recent token compression methods, such as pruning and merging, have shown promise in reducing redundancy, their evaluation remains fragmented and inconsistent. In this work, we present UniPruneBench, a unified and extensible benchmark for visual token pruning in multimodal LLMs. UniPruneBench provides standardized protocols across six ability dimensions and ten datasets, covering ten representative compression algorithms and three families of LMMs (LLaVA-v1.5, Intern-VL3, and Qwen2.5-VL). Beyond task accuracy, it incorporates system-level metrics such as runtime and prefilling latency to provide a holistic view. Our experiments uncover several key findings: (1) random pruning is a surprisingly strong baseline, (2) no single method consistently outperforms others across scenarios, (3) pruning sensitivity varies significantly across tasks, with OCR being most vulnerable, and (4) pruning ratio is the dominant factor governing performance degradation. We believe UniPruneBench will serve as a reliable foundation for future research on efficient multimodal modeling.
ACM International Conference on Multimedia (ACM MM) 2026 · 14 citations
arXiv (Cornell University) 2026 · 0 citations
Last synced
This work has 18 recorded citations, but citing papers have not been linked locally yet.
No comments yet — start the discussion below.