Papers › VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs

VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs

11 Jun 2024arXiv:2406.07476archive 2025-07-28

Zesen Cheng, Sicong Leng, Hang Zhang, Yifei Xin, Xin Li, Guanzheng Chen, Yongxin Zhu, Wenqi Zhang, Ziyang Luo, Deli Zhao, Lidong Bing

In this paper, we present the VideoLLaMA 2, a set of Video Large Language Models (Video-LLMs) designed to enhance spatial-temporal modeling and audio understanding in video and audio-oriented tasks. Building upon its predecessor, VideoLLaMA 2 incorporates a tailor-made Spatial-Temporal Convolution (STC) connector, which effectively captures the intricate spatial and temporal dynamics of video data. Additionally, we integrate an Audio Branch into the model through joint training, thereby enriching the multimodal understanding capabilities of the model by seamlessly incorporating audio cues. Comprehensive evaluations on multiple-choice video question answering (MC-VQA), open-ended video question answering (OE-VQA), and video captioning (VC) tasks demonstrate that VideoLLaMA 2 consistently achieves competitive results among open-source models and even gets close to some proprietary models on several benchmarks. Furthermore, VideoLLaMA 2 exhibits reasonable improvements in audio-only and audio-video question-answering (AQA & OE-AVQA) benchmarks over existing models. These advancements underline VideoLLaMA 2's superior performance in multimodal comprehension, setting a new standard for intelligent video analysis systems. All models are public to facilitate further research.

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="2406.07476")

Code

Syntology Ran 7 of 17 code samples harvested from 1 repository linked to this paper; 10 have no recorded run. Of those that ran: 1 ran · violated contract; 4 ran · our draft was wrong; 1 ran · fixture could not drive it; 1 ran with no contract checked.

By repository: community (archive-listed): 17 samples from 1 repository, 7 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

damo-nlp-sg/videollama2 officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
damo-nlp-sg/inf-clip mentioned on GitHubpytorchApache-2.0 report
damo-nlp-sg/videollama3 mentioned on GitHubpytorchApache-2.0 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

17 samples harvested; 7 ran; 0 honoured the contract we drafted; 10 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.

1ran · violated contract
4ran · our draft was wrong
1ran · fixture could not drive it
1ran
10unverified

Licence: 0 of the 17 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 damo-nlp-sg/inf-clip. “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.

basic_clean damo-nlp-sg/inf-clip/inf_clip/models/tokenizer.py community (archive-listed) ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 98f385d847636a3e · report
convert_to_custom_text_state_dict damo-nlp-sg/inf-clip/inf_clip/models/clip_arch.py community (archive-listed) ran · violated contract Apache-2.0 (permissive) · f19962ebb134b3d7 · report
get_1d_sincos_pos_embed_from_grid damo-nlp-sg/inf-clip/inf_clip/models/pos_embed.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 4bdd36eab04c3f62 · report
get_cast_dtype damo-nlp-sg/inf-clip/inf_clip/models/clip_arch.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · dcd422d66b0581d8 · report
get_input_dtype damo-nlp-sg/inf-clip/inf_clip/models/clip_arch.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · b476c8cfbf0f1f47 · report
get_pairs damo-nlp-sg/inf-clip/inf_clip/models/tokenizer.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · d919ae32e5e4e616 · report
load_state_dict damo-nlp-sg/inf-clip/inf_clip/factory.py community (archive-listed) ran Apache-2.0 (permissive) · 0b94bc3f6eb369a4 · report
all_reduce damo-nlp-sg/inf-clip/inf_clip/models/loss.py community (archive-listed) unverified Apache-2.0 (permissive) · 86e38328c37d09a8 · report
gather_features damo-nlp-sg/inf-clip/inf_clip/models/loss.py community (archive-listed) unverified Apache-2.0 (permissive) · 9ad02daca617f28b · report
get_2d_sincos_pos_embed damo-nlp-sg/inf-clip/inf_clip/models/pos_embed.py community (archive-listed) unverified Apache-2.0 (permissive) · 9417ae492629cf06 · report
get_2d_sincos_pos_embed_from_grid damo-nlp-sg/inf-clip/inf_clip/models/pos_embed.py community (archive-listed) unverified Apache-2.0 (permissive) · 4fd80de79832745d · report
get_model_config damo-nlp-sg/inf-clip/inf_clip/factory.py community (archive-listed) unverified Apache-2.0 (permissive) · 5c7cf820e99443ff · report
get_tokenizer damo-nlp-sg/inf-clip/inf_clip/factory.py community (archive-listed) unverified Apache-2.0 (permissive) · cd2f59a566701fe4 · report
neighbour_exchange damo-nlp-sg/inf-clip/inf_clip/models/loss.py community (archive-listed) unverified Apache-2.0 (permissive) · e332856e3c2fc814 · report
prepare_inputs_for_generation damo-nlp-sg/inf-clip/inf_clip/models/coca_arch.py community (archive-listed) unverified Apache-2.0 (permissive) · fb651d0a97fd4d3f · report
register_pooler damo-nlp-sg/inf-clip/inf_clip/models/hf_model.py community (archive-listed) unverified Apache-2.0 (permissive) · 2a377da4a76a2d44 · report
whitespace_clean damo-nlp-sg/inf-clip/inf_clip/models/tokenizer.py community (archive-listed) unverified Apache-2.0 (permissive) · 62c3f3fab07e9ee3 · report

Tasks

Multiple-choiceQuestion AnsweringTemporal Relation ExtractionVideo CaptioningVideo 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 VideoLLaMA2-72B Group Score 8.4 #8 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground VideoLLaMA2-72B Text Score 36.2 #8 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground VideoLLaMA2-72B Video Score 21.8 #8 of 24 Archive leaderboard report
Video Question Answering MVBench VideoLLaMA2 (72B) Avg. 62.0 #6 of 22 Archive leaderboard report
Video Question Answering NExT-QA VideoLLaMA2.1(7B) Accuracy 75.6 #23 of 47 Archive leaderboard report
Video Question Answering Perception Test VideoLLaMA2 (72B) Accuracy (Top-1) 57.5 #4 of 6 Archive leaderboard report
Video Question Answering TVBench VideoLLaMA2 72B Average Accuracy 48.4 #14 of 28 Archive leaderboard report
Video Question Answering TVBench VideoLLaMA2 7B Average Accuracy 42.9 #19 of 28 Archive leaderboard report
Video Question Answering TVBench VideoLLaMA2.1 Average Accuracy 42.1 #22 of 28 Archive leaderboard report
Zero-Shot Video Question Answer EgoSchema (fullset) VideoLLaMA2 (72B) Accuracy 63.9 #6 of 29 Archive leaderboard report
Zero-Shot Video Question Answer VNBench VideoLLaMA2 Accuracy 4.5 #8 of 9 Archive leaderboard report
Zero-Shot Video Question Answer Video-MME VideoLLaMA2 (72B) Accuracy (%) 63.1 #9 of 11 Archive leaderboard report
Zero-Shot Video Question Answer Video-MME (w/o subs) VideoLLaMA2 (72B) Accuracy (%) 60.9 #8 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.

Methods

ConvolutionSET

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