Papers › Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

16 Nov 2023arXiv:2311.10122archive 2025-07-28

Bin Lin, Yang Ye, Bin Zhu, Jiaxi Cui, Munan Ning, Peng Jin, Li Yuan

The Large Vision-Language Model (LVLM) has enhanced the performance of various downstream tasks in visual-language understanding. Most existing approaches encode images and videos into separate feature spaces, which are then fed as inputs to large language models. However, due to the lack of unified tokenization for images and videos, namely misalignment before projection, it becomes challenging for a Large Language Model (LLM) to learn multi-modal interactions from several poor projection layers. In this work, we unify visual representation into the language feature space to advance the foundational LLM towards a unified LVLM. As a result, we establish a simple but robust LVLM baseline, Video-LLaVA, which learns from a mixed dataset of images and videos, mutually enhancing each other. Video-LLaVA achieves superior performances on a broad range of 9 image benchmarks across 5 image question-answering datasets and 4 image benchmark toolkits. Additionally, our Video-LLaVA also outperforms Video-ChatGPT by 5.8%, 9.9%, 18.6%, and 10.1% on MSRVTT, MSVD, TGIF, and ActivityNet, respectively. Notably, extensive experiments demonstrate that Video-LLaVA mutually benefits images and videos within a unified visual representation, outperforming models designed specifically for images or videos. We aim for this work to provide modest insights into the multi-modal inputs for the LLM. Code address: https://github.com/PKU-YuanGroup/Video-LLaVA

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

No sample from the archive's official repository (PKU-YuanGroup/Video-LLaVA) is linked to this paper; the 6 samples below come from 2 community repositories, and 1 more is identical to code Syntology harvested from another paper's repository. Syntology harvested code from PKU-YuanGroup/Video-LLaVA for another paper.

PKU-YuanGroup/Video-LLaVA: Syntology's graph links these samples to arXiv:2507.15504 (its page on Syntology; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 3 honoured, 1 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method.

Syntology 7 code samples: 6 from 2 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 7: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 6 samples from 2 repositories, 5 ran (5 with no instrument failure; 0 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

PKU-YuanGroup/Video-LLaVA officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to arXiv:2507.15504 (its page on Syntology; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 3 honoured, 1 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
PKU-YuanGroup/LLMBind mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
PKU-YuanGroup/MoE-LLaVA mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Visual Instruction Tuning (arXiv:2304.08485; that paper's own run record: 16 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 8 where Syntology's instrument failed) · 35 unverified); nothing checks that they implement this paper's method. report
pku-yuangroup/languagebind mentioned on GitHubpytorchMITSyntology: 5 samples, 5 ran (0 instrument-failed) · 0 unverified report
pku-yuangroup/video-bench mentioned on GitHubSyntology: not harvested report
qiujihao19/artemis mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interest (arXiv:2307.03601; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

7 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
3ran
2unverified

Licence: 2 of the 7 samples are pointer only, meaning Syntology does not serve that copy's text. For 1 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 1 sample is listed last, in a group without a repository. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

PKU-YuanGroup/LLMBind community (archive-listed) 1 sample, 0 ran · 1 unverified
LlavaMetaForCausalLM in 16 papers model/llava/model/llava_arch.py unverified Apache-2.0 (permissive) · 3d9df14425d0ebcf · report
pku-yuangroup/languagebind community (archive-listed) 5 samples, 5 ran (0 instrument-failed) · 0 unverified
get_depth_transform languagebind/depth/processing_depth.py ran MIT (permissive) · ac6a026c5d4ff47b · report fixture: signature config:MockConfig/0/? · returned Compose · level recorded 2026-09-24
int16_to_float32_torch in 5 papers languagebind/audio/processing_audio.py ran · honoured contract fingerprinted MIT (permissive) · 20a58a8c04ccda43 · report fixture: signature x:Tensor/2/int · returned Tensor · level recorded 2026-09-22
make_list_of_images languagebind/audio/processing_audio.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 852e87ab903330a7 · report fixture: signature x:Tensor/2/float · returned list · level recorded 2026-09-25
opencv_loader languagebind/depth/processing_depth.py ran MIT (permissive) · 5530423121ac4757 · report fixture: signature path:str/0/str · returned ndarray · level recorded 2026-09-24
stack_dict model/languagebind.py ran MIT (permissive) · 833effea5806a36e · report fixture: signature x:list/1/dict, device:device/0/? · returned dict · level recorded 2026-09-24
Repository not located 1 sample, 0 ran · 1 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
build_video_tower repository not located unverified licence of this copy not recorded · 6d6b281af73773f5 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for Video-LLaVA (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Video-LLaVA (graph method: Video-LLaVA). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. build_video_tower entry · level 0 · unverified · 6d6b281af73773f5

harvested Assembly for Video-LLaVA (graph method: Video-LLaVA). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. LlavaMetaForCausalLM entry · level 0 · unverified · 3d9df14425d0ebcf

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2311.10122") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModelingLanguage ModellingLarge Language ModelMultiple-choiceQuestion AnsweringTemporal Relation ExtractionVideo Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)Zero-Shot Video Question Answer

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Temporal Relation Extraction Vinoground Video-LLaVA-7B Group Score 6.6 #15 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Video-LLaVA-7B Text Score 24.8 #15 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Video-LLaVA-7B Video Score 25.8 #15 of 24 Archive leaderboard report
Video Question Answering ActivityNet-QA Video-LLaVA Accuracy 45.3 #16 of 36 Archive leaderboard report
Video Question Answering ActivityNet-QA Video-LLaVA Confidence score 3.3 #16 of 36 Archive leaderboard report
Visual Question Answering MM-Vet Video-LLaVA GPT-4 score 32.0 #187 of 231 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA Video-LLaVA Accuracy 45.3 #22 of 28 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA Video-LLaVA Confidence Score 3.3 #22 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA Video-LLaVA-7B Accuracy 59.2 #17 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA Video-LLaVA-7B Confidence Score 3.5 #17 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA Video-LLaVA-7B Accuracy 70.7 #17 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA Video-LLaVA-7B Confidence Score 3.9 #17 of 28 Archive leaderboard report
Zero-Shot Video Question Answer TGIF-QA Video-LLaVA-7B Accuracy 70.0 #9 of 14 Archive leaderboard report
Zero-Shot Video Question Answer TGIF-QA Video-LLaVA-7B Confidence Score 4.0 #9 of 14 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