AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

HTTM: Head-wise Temporal Token Merging for Faster VGGT

Weitian Wang, Lukas Meiner, Rai Shubham, Cecilia De La Parra, Akash Kumar

CVPR 2026 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add adaptive outlier filtering (10%) to block-wise merging (ablation)HTTM · NRGBDaccuracy (Chamfer): 0.240 -> 0.012Improved
5% vs 10% outlier filtering (ablation)HTTM · NRGBDaccuracy (Chamfer): 0.013 vs 0.012No clear effect
apply HTTM to VGGT (Q 0.20, K/V 0.30)VGGT · 7 Scenes (stride 10)accuracy (Chamfer): 0.019 -> 0.020Hurt
apply HTTM to VGGTVGGT · 7 Scenes (stride 10)time: 9.1s -> 4.3sImproved
apply HTTM to VGGTVGGT · NRGBD (stride 10)accuracy (Chamfer): 0.010 -> 0.012Hurt
HTTM vs FastVGGTVGGT · NRGBD (stride 10)accuracy (Chamfer): 0.016 -> 0.012Improved
HTTM vs FastVGGTVGGT · ETH3Daccuracy (Chamfer): 0.120 -> 0.091Improved
HTTM vs FastVGGTVGGT · VGGT latency, 1000 frameslatency: 175.2s -> 102.8sImproved
block-wise merging after temporal reordering vs global matchingHTTM vs FastVGGT · Global attention latency, 1000 framesmatching time: 2.31s -> 0.12sImproved
adaptive outlier filtering increases aggregation overheadHTTM vs FastVGGT · Global attention latency, 1000 framesmerged token aggregation time: 0.11s -> 0.41sHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy