The paper argues that token importance alone is insufficient to determine safe removal of visual tokens in multimodal large language models, because removability depends on representation depth and the surrounding deletion set. Through controlled experiments, the authors show that the same tokens can have different effects when removed at different depths or contexts. They introduce CoRePrune, a training‑free two‑stage pruning framework that refreshes deletion effects as visual representations evolve and refines candidate tokens based on the current deletion set, achieving high performance retention across multiple backbones and reducing prefill time significantly.
By Shengli He, Yongchao Liang, Roumeng He, Junjie Zeng, Jiyuan He, Xin Fang, Can Wu, Li Zheng
arXiv:2609.32353v2 Announce Type: replace
Abstract: Visual token reduction is an effective way to accelerate multimodal large language models (MLLMs), but performance deteriorates rapidly under extre...
By Junxian Li, Ruixuan Yang, Tianao Zhang, Tiange Xu, Weisheng Dong, Yulun Zhang
arXiv:2608. 07585v1 Announce Type: cross Abstract: Long-video understanding requires models to efficiently acquire and reuse sparse visual evidence from long and redundant video streams.
By Zijian Wang, Junnan Zhu, Rongzhen Li, Xiao Liu, Guohui Xiang, Quan Lu, Lijia Liu, Yining Wang, Jiang Zhong, Kaiwen Wei
arXiv:2606. 25232v1 Announce Type: new Abstract: Ordered bottlenecks aim to provide utility at flexible budgets by assigning coarse information to early tokens and task-relevant detail to later ones.
By Erik Ayari, Manuel Traub, Martin V. Butz
ScreenSearch is a system for exploring desktop GUI states under partial observability. It combines structural screen retrieval, deduplication, and an ambiguity-aware PUCT graph-bandit to expand the reachable frontier while reducing state ambiguity. Across 11 applications, it collected over 1 M screenshots and 30 K deduplicated states, demonstrating that both ambiguity reduction and frontier exploration are essential for effective corpus building.
By Michael Solodko, Justin Wagle
VisCache introduces a two-stage, plug‑and‑play framework for pruning visual key‑value caches in Vision Large Language Models without retraining. The first stage filters out temporally redundant keyframes, while the second stage, PruneKV, applies a parabolic layer‑wise budget and asymmetric update to selectively prune keys and fuse values, preserving essential context. Experiments show up to 2.35× speedup and significant memory savings with only 19–28% of the original cache retained, outperforming existing baselines.
By Lyuke Wang, Zhuo Li, Guangxu Zhu
arXiv:2606. 01790v1 Announce Type: cross Abstract: Vision-language-model-based graphical user interface (GUI) agents have shown broad automation capabilities, yet deployment is bottlenecked by a key-value (KV) cache that grows linearly with interaction steps.
By Yuhang Han, Wenzheng Yang, Yujie Chen, Xiangqi Jin, Yaojie Zhang, Siteng Huang, Linfeng Zhang
The paper introduces Interaction‑Aligned Pruning (IAprune), a training‑free method for visual token pruning in embodied manipulation tasks. IAprune jointly decides per‑frame budget and token selection, using semantic‑motion spatial agreement to choose between conservative and aggressive coverage, and applies geometric residual correction to focus on under‑represented boundaries. Experiments on four policies, three simulation benchmarks, and a real‑robot platform show that IAprune matches unpruned performance on LIBERO while achieving up to 1.54× speed‑up and 1.48× acceleration on a real robot.
By Jintao Cheng, Weibin Li, Haozhe Wang, Gang Wang, Yipu Zhang, Xiaoyu Tang, Jin Wu, Xieyuanli Chen, Yunhui Liu, Wei Zhang
Recent advancements in MLLM-based long-form video understanding have mitigated inference-time computational cost and limited context lengths by selecting query-relevant frames. However, existing approaches predominantly rely on external proxy scorers and rigid heuristic rules, inevitably suffering from misalignment with the target MLLM's intrinsic evidence and failing to accommodate the non-uniform spatiotemporal information density.
arXiv:2609.39547v1 Announce Type: new
Abstract: GUI agents built on large language and vision-language models still struggle on unseen applications and complex multi-step tasks, as completing real GU...
By Bo Han, Qianyi Wang, Shuai Liu, Xiong Zifan, Changqiao Wu, Yuanfa Li, Pengzhi Gao, Wei Liu, Jian Luan, Heng Qu, Yunpeng Song, Zhongmin Cai
arXiv:2608.29897v1 Announce Type: new
Abstract: Long-horizon agents need a context manager to compress growing interaction histories into a bounded working context, via passive strategies or active s...
By Jiaqi Su, Cong Pang, Jiawei Hong, Tiankuo Yao, Zixuan Chen, Xin Lou, Lewei Lu
arXiv:2608. 09176v1 Announce Type: cross Abstract: Visual token compression for vision--language models (VLMs) has largely relied on criteria such as attention, redundancy, and uncertainty to maximize average accuracy under a fixed compute budget, implicitly assuming that all errors carry equal cost.
By Jingbo Wen, Liang He, Mingyu Cao, Haoyu Wang, Minxuan Hu, Kangning Cui, Xilu Wang