AI Paper Atlas

Research map / Video / Video-language understanding

NaQ: Leveraging Narrations As Queries To Supervise Episodic Memory

Santhosh Kumar Ramakrishnan, Ziad Al-Halah, Kristen Grauman

CVPR 2023 · Video › Video-language understanding · approach: Video-language pretraining · 18 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 narration-derived queries to VSLNet trainingVSLNet · Ego4D NLQ valR@1, IoU=0.3: 5.45 -> 10.26Improved
Add narration-derived queries to EgoVLP trainingEgoVLP · Ego4D NLQ valR@1, IoU=0.3: 10.84 -> 15.90Improved
Add narration-derived queries to ReLER* trainingReLER* · Ego4D NLQ valR@1, IoU=0.3: 14.66 -> 19.31Improved
Add Ego4D narration-derived query trainingVSLNet · TACoS valR@1, IoU=0.3: 20.10 -> 23.39Improved
Add Ego4D narration-derived query trainingEgoVLP · TACoS valR@1, IoU=0.3: 16.52 -> 18.24Improved
Add temporal response jittering to narration windows (ablation)NaQ with VSLNet · Ego4D NLQ valR@1, IoU=0.3: 9.80 -> 10.26Improved
Add temporal response jittering to narration windows (ablation)NaQ with EgoVLP · Ego4D NLQ valR@1, IoU=0.3: 15.25 -> 15.90Improved
Add temporal response jittering to narration windows (ablation)NaQ with ReLER* · Ego4D NLQ valR@1, IoU=0.3: 18.51 -> 19.31Improved
Add NLQ-only finetuning after joint NaQ and NLQ training (ablation)VSLNet + NaQ · Ego4D NLQ valR@1, IoU=0.3: 10.26 -> 9.34Hurt
Use temporal expansion scale S=5.0 rather than S=2.5 (ablation)EgoVLP + NaQ · Ego4D NLQ valR@1, IoU=0.5: 8.64 -> 9.46Improved
Train with NaQ queries for low-shot objectsVSLNet · Ego4D NLQ val, low-shot objectsR@1, IoU=0.3: 4.05 -> 10.30Improved
Train with NaQ queries for query template 'In what location did I see X?'ReLER* · Ego4D NLQ valR@1, IoU=0.5: 5.46 -> 4.78Hurt

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