method · first seen 5 Oct 2026
token pruning
Also: ToMe, patch pruning, token merging, token reduction, token sparsification, bipartite token merging
Removing less relevant tokens during model inference to reduce computation with limited loss in performance.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Less Context, Better Geometry: Masked Geometric Encoder for Robust 3D Foundation ModelsPapers · 5 Oct 2026 · Zhimin Shao, Xijun Liu, Zhaoliang Zhang, Yutao Tang, Abhay Yadav, Rama Chellappa…
- Later Is Better: Token Reduction for ViTs Under Distribution ShiftPapers · 6 Oct 2026 · Hyeongheon Cha, Hyungjun Yoon, Sung-Ju Lee
- DIPrune: Task-Aware Token Pruning with Dual Importance for Efficient Multimodal Language ModelsPapers · 6 Oct 2026 · Shuo Yang, Changbai Li, Linlin Yang, Huobin Tan, Rongyu Chen, Tongfei Chen, Tian…
- ornl-utome/UtoMeCode · 7 Oct 2026 · ornl-utome · 22 stars
- LapMerge: Differentiable Exact-Budget Token Merging in Linear TimePapers · 9 Oct 2026 · Magdalena Trędowicz, Aleksandra Nowak, Tomasz Steifer, Łukasz Struski, Szymon…
- TwinPrune: Density-Aware Two-Phase Token Pruning for Vision-Language ModelsPapers · 13 Oct 2026 · Kai Liu, Anqi Li, Junxian Li, Zhixin Wang, Zhikai Chen, Renjing Pei, Yulun Zhang
Latest sources
- SWAM: Saliency-Weighted Asymmetric Merging for Efficient Vision TransformersPapers · 13 Nov 2026 · Utkarsh Kumar Singh, Manish Das, Shiv Ram Dubey, Satish Kumar Singh
- FoveaDet: Software-Hardware Co-Design of Foveated CNN-Transformer for Object DetectionPapers · 11 Nov 2026 · Deepak Mewada, Apporva Saraf, Ashwin Lele
- Towards Real-Time Full-Waveform LiDAR Transformers via Intensity-Guided Token Reduction and Physics-Aware AugmentationPapers · 5 Nov 2026 · Kotaro Oishi, Kazuma Ikeda, Ryosei Hara, Ryo Yoshida, Mariko Isogawa, Kentaro Yo…
- Training-Free Predictive Spatio-Temporal Token Pruning for Few-Step Autoregressive Video DiffusionPapers · 5 Nov 2026 · Jia Li, Xurui Peng, Xing Wang, Hayden Kwok-Hay So
- Where Is VLA Inference Time Spent? Bottlenecks and Optimization Interactions Across GPUsPapers · 4 Nov 2026 · Muhammad Danish Waseem, Ahmed Ali-Eldin Hassan
- R$^2$-Omni: Reducing Structural and Precision Redundancy in Omni-LLMs for Long-Form GenerationPapers · 2 Nov 2026 · Bolong Feng, Pengfei Zhou, Zhenglin Wan, Wei Wang, Wangbo Zhao
- Calibrating Attention-Based Token Pruning for Efficient Multi-Image Synthesis in UMMsPapers · 29 Oct 2026 · Zimeng Wu, Qijian Tian, Penghui Du, Changqian Yu, Jiaxin Chen, Yunhong Wang
- ConvRoute-ViT: Convolutional Token Routing for Efficient Vision TransformersPapers · 28 Oct 2026 · Manish Das, Shiv Ram Dubey, Satish Kumar Singh
- GPA-VGGT: Geometry-Preserving Acceleration of VGGTPapers · 27 Oct 2026 · Muyeol Sung, Jangwon Kim, Jongwoo Lim
- Training-Free Inference Acceleration for Video Large Language Models: Methods, Trade-offs, and EvaluationPapers · 26 Oct 2026 · Yuqi Li, Chuanyue Yan, Neil Rodrigues, Xianglong Wang, Xiaoqin Feng, Junming Liu…
- ToDoPro: Faster and Better Reference-Based Generation: Early Token Reduction via Downsampled Self-ProbingPapers · 26 Oct 2026 · Caiyi Sun, Yujing Sun, Siu-Ming Yiu
- Consommé: Conditioned Token Merging for Split InferencePapers · 23 Oct 2026 · Man Ngo, Jan S. Rellermeyer