Papers › Learning Video Representations from Large Language Models

Learning Video Representations from Large Language Models

8 Dec 2022CVPR 2023 1arXiv:2212.04501archive 2025-07-28

Yue Zhao, Ishan Misra, Philipp Krähenbühl, Rohit Girdhar

We introduce LaViLa, a new approach to learning video-language representations by leveraging Large Language Models (LLMs). We repurpose pre-trained LLMs to be conditioned on visual input, and finetune them to create automatic video narrators. Our auto-generated narrations offer a number of advantages, including dense coverage of long videos, better temporal synchronization of the visual information and text, and much higher diversity of text. The video-text embedding learned contrastively with these additional auto-generated narrations outperforms the previous state-of-the-art on multiple first-person and third-person video tasks, both in zero-shot and finetuned setups. Most notably, LaViLa obtains an absolute gain of 10.1% on EGTEA classification and 5.9% Epic-Kitchens-100 multi-instance retrieval benchmarks. Furthermore, LaViLa trained with only half the narrations from the Ego4D dataset outperforms baseline models trained on the full set, and shows positive scaling behavior on increasing pre-training data and model size.

PaperPDFConference PDFCodeCode Syntology ran

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

Syntology 20 code samples harvested from 1 repository linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 2 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 20 samples from 1 repository, 13 ran (9 with no instrument failure; 4 where Syntology's instrument failed). 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 2 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. 2 of the 13 samples that ran have no recorded date.

facebookresearch/lavila officialmentioned in papermentioned on GitHubpytorchMITSyntology: 20 samples, 13 ran (4 instrument-failed) · 7 unverified report
Ziyang412/VideoTree mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos (arXiv:2405.19209; that paper's own run record: 11 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 2 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
ceezh/llovi mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to A Simple LLM Framework for Long-Range Video Question-Answering (arXiv:2312.17235; that paper's own run record: 6 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; 3 where Syntology's instrument failed) · 0 unverified) and ACL Anthology id 2024.emnlp-main.1209 (that paper's own run record: 6 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; 3 where Syntology's instrument failed) · 0 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

20 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 2 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · violated contract
3ran · our draft was wrong
1ran; the contract check could not run
7ran
7unverified

Licence: 20 of the 20 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 facebookresearch/lavila. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

facebookresearch/lavila official repository 20 samples, 13 ran (4 instrument-failed) · 7 unverified
attn in 9 papers lavila/models/timesformer.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT recorded; this copy not marked cleared · pointer only · 544c604bc4f58d28 · report fixture: signature q:Tensor/3/float, k:Tensor/3/float, v:Tensor/3/float · returned Tensor · level recorded 2026-09-22
augment_gpt2_config lavila/models/gpt2_gated.py ran MIT recorded; this copy not marked cleared · pointer only · 84ff8738aacf9517 · report fixture: signature config:MockConfig/0/?, cross_attn_freq:int/0/int, gated_xattn:bool/0/bool · returned MockConfig · level recorded 2026-09-25
basic_clean in 217 papers lavila/models/tokenizer.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · 98f385d847636a3e · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-22
convert_to_normal_tensor lavila/models/distributed_utils.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 8e370c62b78a927f · report fixture: signature tensor:Tensor/2/float, orig_device:str/0/str · returned Tensor · level recorded 2026-09-25
default in 430 papers lavila/models/coca.py ran · violated contract fingerprinted MIT recorded; this copy not marked cleared · pointer only · 60fff7c3c400d7ff · report fixture: signature val:int/0/int, d:int/0/int · returned int · level recorded, date not recorded
exists in 528 papers
census: 5 implementations from 103 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 1, kind int, dtype int64
lavila/models/coca.py ran · violated contract MIT recorded; this copy not marked cleared · pointer only · aa5486a3650902d8 · report fixture: signature val:Tensor/1/int · returned bool · level recorded, date not recorded
gather_features in 78 papers lavila/models/loss.py ran MIT recorded; this copy not marked cleared · pointer only · 19519e7e08dd0087 · report fixture: signature image_features:Tensor/2/float, text_features:Tensor/2/float, local_loss:bool/0/bool, gather_with_grad:bool/0/bool, rank:int/0/int, world_size:int/0/int · returned tuple · level recorded 2026-09-25
gather_from_all lavila/models/distributed_utils.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · b57f5d4b094326f8 · report fixture: signature tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-25
get_metric_names in 11 papers lavila/models/models.py ran MIT recorded; this copy not marked cleared · pointer only · 051362c936e3619d · report fixture: signature model:str/0/str · returned list · level recorded 2026-09-25
get_pairs in 260 papers lavila/models/tokenizer.py ran · our draft was wrong MIT recorded; this copy not marked cleared · pointer only · d919ae32e5e4e616 · report fixture: signature word:tuple/1/str · returned set · level recorded 2026-09-22
inflate_positional_embeds lavila/models/utils.py ran MIT recorded; this copy not marked cleared · pointer only · d8e9e992e4668cbb · report fixture: signature current_model_state_dict:dict/0/?, new_state_dict:dict/0/?, num_frames:int/0/int, load_temporal_fix:str/0/str · returned dict · level recorded 2026-09-25
sim_matrix in 12 papers lavila/models/loss.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · da8e97ddc50728ed · report fixture: signature a:Tensor/2/float, b:Tensor/2/float, eps:float/0/float · returned Tensor · level recorded 2026-09-25
whitespace_clean in 230 papers lavila/models/tokenizer.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · 9542161e9640b858 · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-22
build_model in 469 papers lavila/models/openai_model.py unverified MIT recorded; this copy not marked cleared · pointer only · 7144c613fbeb2f92 · report
convert_to_distributed_tensor lavila/models/distributed_utils.py unverified MIT recorded; this copy not marked cleared · pointer only · bcfb2b11cc600666 · report
get_model in 368 papers lavila/utils/distributed.py unverified MIT recorded; this copy not marked cleared · pointer only · 2ec9ac675733a9ac · report
load in 392 papers lavila/models/openai_clip.py unverified MIT recorded; this copy not marked cleared · pointer only · 97742b8486aa64f2 · report
load_tf_weights_in_gpt2 in 68 papers lavila/models/gpt2_gated.py unverified MIT recorded; this copy not marked cleared · pointer only · bc9fb74e483fb901 · report
rgetattr in 22 papers lavila/models/utils.py unverified MIT recorded; this copy not marked cleared · pointer only · 44925cf3831bfe79 · report
rsetattr in 10 papers lavila/models/utils.py unverified MIT recorded; this copy not marked cleared · pointer only · f9de4cdd8a624801 · report

Tasks

Action ClassificationAction RecognitionDiversityEgocentric Activity RecognitionMulti-Instance RetrievalSelf-Supervised Action Recognition Linear

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Recognition Charades-Ego LaViLa (Finetuned, TimeSformer-L) mAP 36.1 #1 of 6 Archive leaderboard report
Action Recognition Charades-Ego LaViLa (Zero-shot, TimeSformer-L) mAP 28.9 #5 of 6 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 LaViLa (TimeSformer-L) Action@1 51 #7 of 32 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 LaViLa (TimeSformer-L) Noun@1 62.9 #7 of 32 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 LaViLa (TimeSformer-L) Verb@1 72 #7 of 32 Archive leaderboard report
Egocentric Activity Recognition EGTEA LaViLa (Finetuned, TimeSformer-L) Average Accuracy 81.75 #1 of 6 Archive leaderboard report
Egocentric Activity Recognition EGTEA LaViLa (Finetuned, TimeSformer-L) Mean class accuracy 76 #1 of 6 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