AI Paper Atlas

Research map / Video / Video-language understanding

Rethinking Video-Language Model from the Language Input Perspective

Xiang Fang, Wanlong Fang, Changshuo Wang, Xiaoye Qu, Daizong Liu

AAAI 2026 · Video › Video-language understanding · approach: Video-language pretraining · 0 citations (OpenAlex)

Read the paper

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
augment query texts (no plug-in)X-Pool · MSR-VTT text-to-videoR@1: 46.9 -> 40.1Hurt
add the plug-and-play frameworkX-Pool with augmented text · MSR-VTT text-to-videoR@1: 40.1 -> 45.9Improved
add the plug-and-play frameworkT-MASS with augmented text · MSR-VTT text-to-videoR@1: 42.1 -> 51.4Improved
add the plug-and-play frameworkT-MASS without augmentation · MSR-VTT text-to-videoR@1: 50.2 -> 52.3Improved
add the plug-and-play frameworkBLIP-2 with augmented text · STAR Interactionaccuracy: 60.9 -> 66.2Improved
add the plug-and-play frameworkG2L with augmented text · ActivityNet Captions R@1 IoU=0.5R@1: 33.01 -> 46.86Improved
add multi-level text augmentation (ablation)Ours (G2L) · ActivityNet Captions R@1 IoU=0.5R@1: 40.28 -> 46.86Improved
add attribute-based text reasoning (ablation)Ours (G2L) · ActivityNet Captions R@1 IoU=0.5R@1: 42.03 -> 46.86Improved
add video-guided self-weighted cross-modal bridging (ablation)Ours (G2L) · ActivityNet Captions R@1 IoU=0.5R@1: 43.86 -> 46.86Improved
use own augmentation instead of back-translation (ablation)Ours (BLIP-2) · NExT-QA Temporalaccuracy: 62.8 -> 69.2Improved
use own augmentation instead of LLaMA-7B (ablation)Ours (BLIP-2) · NExT-QA Temporalaccuracy: 66.8 -> 69.2Improved
add S1 significance weight (ablation)Ours (BLIP-2) · STAR Interactionaccuracy: 64.8 -> 66.2Improved

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