Papers › An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM

An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM

27 Mar 2024arXiv:2403.18406archive 2025-07-28

Wonkyun Kim, Changin Choi, Wonseok Lee, Wonjong Rhee

Stimulated by the sophisticated reasoning capabilities of recent Large Language Models (LLMs), a variety of strategies for bridging video modality have been devised. A prominent strategy involves Video Language Models (VideoLMs), which train a learnable interface with video data to connect advanced vision encoders with LLMs. Recently, an alternative strategy has surfaced, employing readily available foundation models, such as VideoLMs and LLMs, across multiple stages for modality bridging. In this study, we introduce a simple yet novel strategy where only a single Vision Language Model (VLM) is utilized. Our starting point is the plain insight that a video comprises a series of images, or frames, interwoven with temporal information. The essence of video comprehension lies in adeptly managing the temporal aspects along with the spatial details of each frame. Initially, we transform a video into a single composite image by arranging multiple frames in a grid layout. The resulting single image is termed as an image grid. This format, while maintaining the appearance of a solitary image, effectively retains temporal information within the grid structure. Therefore, the image grid approach enables direct application of a single high-performance VLM without necessitating any video-data training. Our extensive experimental analysis across ten zero-shot video question answering benchmarks, including five open-ended and five multiple-choice benchmarks, reveals that the proposed Image Grid Vision Language Model (IG-VLM) surpasses the existing methods in nine out of ten benchmarks.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2403.18406")

Code

Syntology Ran 3 of 3 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 3 ran · our draft was wrong.

By repository: official repository: 3 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

imagegridworth/IG-VLM officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

3 samples harvested; 3 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

3ran · our draft was wrong

Licence: 0 of the 3 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from imagegridworth/IG-VLM. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

get_llava_and_prompt imagegridworth/IG-VLM/eval_llava_openended.py official repository ran · our draft was wrong fingerprinted BSD-3-Clause (permissive) · 27990140245e0f68 · report
validate_llm_size imagegridworth/IG-VLM/eval_llava_openended.py official repository ran · our draft was wrong BSD-3-Clause (permissive) · c3ad8f46bb951fab · report
validate_video_path imagegridworth/IG-VLM/eval_llava_openended.py official repository ran · our draft was wrong BSD-3-Clause (permissive) · 5f5e72c2cae176b2 · report

Tasks

Language ModelingLanguage ModellingMultiple-choiceQuestion AnsweringVideo Question AnsweringVideo-based Generative Performance BenchmarkingZero-Shot Video Question Answer

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v Consistency 3.13 #8 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v Contextual Understanding 3.61 #8 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v Correctness of Information 3.40 #8 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v Detail Orientation 2.80 #8 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v Temporal Understanding 2.89 #8 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct IG-VLM-GPT4v mean 3.17 #8 of 23 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA IG-VLM Accuracy 58.4 #7 of 28 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA IG-VLM Confidence Score 3.5 #7 of 28 Archive leaderboard report
Zero-Shot Video Question Answer IntentQA IG-VLM Accuracy 65.3 #6 of 13 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA IG-VLM Accuracy 63.8 #9 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA IG-VLM Confidence Score 3.5 #9 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA IG-VLM-34B Accuracy 79.6 #7 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA IG-VLM-34B Confidence Score 4.1 #7 of 28 Archive leaderboard report
Zero-Shot Video Question Answer NExT-QA IG-VLM(LLaVA v1.6) Accuracy 70.9 #9 of 27 Archive leaderboard report
Zero-Shot Video Question Answer NExT-QA IG-VLM (GPT-4) Accuracy 68.6 #12 of 27 Archive leaderboard report
Zero-Shot Video Question Answer TGIF-QA IG-VLM Accuracy 79.1 #6 of 14 Archive leaderboard report
Zero-Shot Video Question Answer TGIF-QA IG-VLM Confidence Score 4.2 #6 of 14 Archive leaderboard report
Zero-Shot Video Question Answer TVQA IG-VLM (no speech, GPT-4V) Accuracy 57.8 #2 of 9 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections