AI Paper Atlas

Research map / Video / Video-language understanding

LGDN: Language-Guided Denoising Network for Video-Language Modeling

Haoyu Lu, Mingyu Ding, Nanyi Fei, Yuqi Huo, Zhiwu Lu

NeurIPS 2022 · Video › Video-language understanding · approach: Video-language pretraining · 5 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
Use language-guided SFP instead of selecting frames without SFP (ablation)LGDN · MSR-VTT 1k-A testR@SUM: 336.1 -> 352.9Improved
Add MFCL with MSL instead of standard frame-level contrastive learning (ablation)LGDN · MSR-VTT 1k-A testR@SUM: 330.5 -> 335.2Improved
Add MFCL without MSL (ablation)LGDN without SFP · MSR-VTT 1k-A testR@SUM: 331.2 -> 330.5Hurt
Add MVCL video-level contrastive learning (ablation)LGDN with LSFM and MFCL, without SFP · MSR-VTT 1k-A testR@SUM: 335.2 -> 336.1Improved
Use ensemble global and local alignment at inference instead of local alignment alone (ablation)LGDN · MSR-VTT 1k-A testR@SUM: 352.9 -> 360.4Improved
Use global alignment only at inference instead of local alignment (ablation)LGDN · MSR-VTT 1k-A testR@SUM: 352.9 -> 330.7Hurt
Combine Momentum and CrossMom relevance scores (ablation)LGDN SFP relevance estimator · MSR-VTT 1k-A testR@SUM: 359.7 -> 360.4Improved
Use momentum encoder supervision for relevance scoring instead of SimDot (ablation)LGDN SFP relevance estimator · MSR-VTT 1k-A testR@SUM: 356.5 -> 359.5Improved
Increase memory bank size from 1,200 to 9,600 (ablation)LGDN · MSR-VTT 1k-A testText-to-video R@1: 36.0 -> 38.9Improved
Increase memory bank size from 9,600 to 19,200 (ablation)LGDN · MSR-VTT 1k-A testText-to-video R@1: 38.9 -> 38.3Hurt
Add SFP frame selectionClip4CLIP · MSR-VTT 1k-A testText-to-video R@1: 44.9 -> 45.3Improved
Use salient sampling with 2 selected frames instead of sparse sampling (ablation)LGDN · DiDeMo testR@SUM: 183.0 -> 198.3Improved

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