AI Paper Atlas

Research map / Video / Video segmentation and tracking

RelayFormer: A Unified Local-Global Attention Framework for Scalable Image and Video Manipulation Localization

Wen Huang, Jiarui Yang, Tao Dai, Jiawei Li, Shaoxiong Zhan, Bin Wang, Shu-Tao Xia

ICLR 2026 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 1 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 GLRA with 1 GLR token (ablation)Relay-ViT · image manipulation avg F1F1: 0.454 -> 0.521Improved
add query-based mask decoder (n=1) (ablation)Relay-ViT · image manipulation avg F1F1: 0.521 -> 0.532Improved
use 2 GLR tokens instead of 1 (ablation)Relay-ViT · image manipulation avg F1F1: 0.532 -> 0.554Improved
use 3 GLR tokens instead of 2 (ablation)Relay-ViT · image manipulation avg F1F1: 0.554 -> 0.524Hurt
add 4D RoPE to GLRA (ablation)Relay-ViT · image manipulation avg F1F1: 0.517 -> 0.554Improved
add temporal GLRA on top of spatial GLRA (ablation)Relay · MOSE100F1: 0.6745 -> 0.6877Improved
add spatial GLRA (vs none) (ablation)Relay · MOSE100F1: 0.6124 -> 0.6745Improved
avoid resizing 4K inputs to 1024x1024 (ablation)Relay · IMD2020F1: 0.350 -> 0.453Improved
use 512 sub-images instead of 256 (with overlap) (ablation)Relay-ViT · image manipulation avg F1F1: 0.477 -> 0.554Improved
add overlap between 256 sub-images (ablation)Relay-ViT · image manipulation avg F1F1: 0.469 -> 0.477Improved
apply GLRA in all layers instead of sparse layers (ablation)Relay-ViT · CASIAv1F1 (%): 73.24 -> 75.50Improved
add video forgery data to image training (ablation)Relay-Seg · image benchmarksF1: COV 0.551 -> 0.569; NIST16 0.335 -> 0.282No clear effect

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