AI Paper Atlas

Research map / Video / Video-language understanding

Large-scale Pre-training for Grounded Video Caption Generation

Evangelos Kazakos, Cordelia Schmid, Josef Sivic

ICCV 2025 · Video › Video-language understanding · approach: Encoder-decoder video captioning · 3 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
pre-train on HowToGround1M then fine-tune on iGround, versus fine-tuning only (ablation)GROVE · iGround test (all frames)AP50: 15.8 -> 40.0Improved
add fine-tuning on iGround after pre-training (ablation)GROVE · iGround test (all frames)CIDEr: 50.6 -> 83.5Improved
train GROVE on automatic annotations instead of applying the annotation pipeline directlyGROVE PT · iGround test (all frames)CIDEr: 40.0 -> 50.6Improved
use GROVE video model instead of still-image GLaMM on center frameGROVE PT · iGround test (center frame)AP50: 20.8 -> 27.0Improved
add large-scale pre-training before fine-tuning on VidSTG (ablation)GROVE · VidSTG test declarativem_sIoU: 61.3 -> 63.7Improved
add large-scale pre-training before fine-tuning on ActivityNet-Entities (ablation)GROVE · ActivityNet-Entities valF1_all: 09.51 -> 13.39Improved
scale pre-training set from 1k to 1M videos (ablation)GROVE · iGround valCIDEr, AP50, recallImproved
add spatio-temporal adapters (ablation)GROVE · iGround val (PT+FT)CIDEr: 88.9 -> 92.6Improved
unfreeze bounding box decoder and projection layers (ablation)GROVE · iGround val (PT+FT)CIDEr: 82.2 -> 92.6Improved
raise temporal objectness threshold from 0.0 (no head) to 0.5 (ablation)GROVE · iGround valAP50Improved
replace Stage 1 image grounded captioner with GIT+Llama-3+OWLv2 (ablation)Automatic annotation · iGround valaverage: 22.5 -> 20.9Hurt
replace SVO triplets with full captions in Stage 2 (ablation)Automatic annotation · iGround valaverage: 22.5 -> 21.5Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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