AI Paper Atlas

Research map / Video / Video-language understanding

ViTAL‑X: Video-Text Alignment with Cross‑Modal Temporal Edits

Sethuraman T V, Savya Khosla, Onkar Susladkar, Aditi Tiwari, Seoung Wug Oh, Kushal Kafle, Joon-Young Lee, Derek Hoiem, Simon Jenni

ECCV 2026 · Video › Video-language understanding · approach: Video-language pretraining · 0 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 sequence composition edit (ablation)ViTAL-X · VideoCompaccuracy: 61.4 -> 67.8Improved
add sequence composition edit (ablation)ViTAL-X · RTimeaccuracy: 59.1 -> 65.3Improved
add clip reordering edit (ablation)ViTAL-X · YouCook2accuracy: 46.5 -> 50.4Improved
add temporal cropping edit (ablation)ViTAL-X · DiDeMoaccuracy: 52.5 -> 55.8Improved
add reversal edit (ablation)ViTAL-X · Avg-Tempaccuracy: 58.4 -> 61.1Improved
add textual hard negatives (ablation)ViTAL-X · Avg-Tempaccuracy: 58.8 -> 61.1Improved
use spatio-temporal adapter instead of temporal-only (frozen backbone) (ablation)ViTAL-X · Avg-Tempaccuracy: +1.8 Avg-TempImproved
add margin temporal loss to contrastive loss (ablation)ViTAL-X · XTE-Benchaccuracy: 69.4 with dual objectiveImproved
use GPT-4o-mini instead of GPT-3.5-Turbo for counterfactual captions (ablation)ViTAL-X · XTE-Benchaccuracy: 67.1 -> 69.4Improved
use open-source LLaMA-3-8B instead of GPT-4o-mini (ablation)ViTAL-X · XTE-Benchaccuracy: 69.4 -> 66.3No clear effect
vary filtering thresholds (3x3 sweep) (ablation)ViTAL-X · Avg-Tempaccuracy: +/-0.4 around defaultNo clear effect
scale to 7B multimodal LLMs without temporal supervisionQwen2.5 / LLaVA-NeXT-Video-7B · XTE-Benchaccuracy: hovering near 56%No 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