Papers › ST-LLM: Large Language Models Are Effective Temporal Learners

ST-LLM: Large Language Models Are Effective Temporal Learners

30 Mar 2024arXiv:2404.00308archive 2025-07-28

Ruyang Liu, Chen Li, Haoran Tang, Yixiao Ge, Ying Shan, Ge Li

Large Language Models (LLMs) have showcased impressive capabilities in text comprehension and generation, prompting research efforts towards video LLMs to facilitate human-AI interaction at the video level. However, how to effectively encode and understand videos in video-based dialogue systems remains to be solved. In this paper, we investigate a straightforward yet unexplored question: Can we feed all spatial-temporal tokens into the LLM, thus delegating the task of video sequence modeling to the LLMs? Surprisingly, this simple approach yields significant improvements in video understanding. Based upon this, we propose ST-LLM, an effective video-LLM baseline with Spatial-Temporal sequence modeling inside LLM. Furthermore, to address the overhead and stability issues introduced by uncompressed video tokens within LLMs, we develop a dynamic masking strategy with tailor-made training objectives. For particularly long videos, we have also designed a global-local input module to balance efficiency and effectiveness. Consequently, we harness LLM for proficient spatial-temporal modeling, while upholding efficiency and stability. Extensive experimental results attest to the effectiveness of our method. Through a more concise model and training pipeline, ST-LLM establishes a new state-of-the-art result on VideoChatGPT-Bench and MVBench. Codes have been available at https://github.com/TencentARC/ST-LLM.

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 11 code samples: 8 from 1 repository linked to this paper, 3 identical to code harvested from another paper's repository. Across all 11: 7 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 8 samples from 1 repository, 6 ran (3 with no instrument failure; 3 where Syntology's instrument failed); 3 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.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 4 of the 7 samples that ran have no recorded date.

TencentARC/ST-LLM officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 8 samples, 6 ran (3 instrument-failed) · 2 unverified 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

11 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 4 where Syntology's instrument failed) · 4 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.

1ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
2ran; the contract check could not run
1ran
4unverified

Licence: 7 of the 11 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 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 TencentARC/ST-LLM. 3 samples are 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. 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.

TencentARC/ST-LLM official repository 8 samples, 6 ran (3 instrument-failed) · 2 unverified
RandomMaskingGenerator stllm/models/utils.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 8bd2d829faa2f7b2 · report fixture: signature num_patches:int/0/int, mask_ratio:float/0/float, batch:int/0/int, device:str/0/str · returned Tensor · level recorded 2026-09-24
apply_rotary_pos_emb in 393 papers
census: 14 implementations from 15 papers, 8 distinct outputs across 7 buckets, one shared input per bucket
stllm/models/modeling_llama_mem.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant Apache-2.0 (permissive) · 9b4dff79d5e6102c · report fixture: signature q:Tensor/4/float, k:Tensor/4/float, cos:Tensor/4/float, sin:Tensor/4/float, position_ids:Tensor/2/int · returned tuple · level recorded, date not recorded
disabled_train in 172 papers stllm/models/blip2.py ran · violated contract Apache-2.0 (permissive) · 4cb732f513d69dfd · report fixture: signature self:callable/0/?, mode:bool/0/bool · returned SimpleModel · level recorded, date not recorded
get_sinusoid_encoding_table in 51 papers stllm/models/utils.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · da651e3979a18f84 · report fixture: signature n_position:int/0/int, d_hid:int/0/int · returned Tensor · level recorded 2026-09-22
rotate_half in 424 papers
census: 24 implementations from 96 papers, 4 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
stllm/models/modeling_llama_mem.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · b99eea6376d1e212 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
tile in 109 papers
census: 5 implementations from 7 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
stllm/models/base_model.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · cb33571427334815 · report fixture: signature x:Tensor/4/float, dim:int/0/int, n_tile:int/0/int · returned Tensor · level recorded 2026-09-24
all_gather_with_grad in 93 papers stllm/models/base_model.py unverified Apache-2.0 (permissive) · 0ec9fc2025c16f65 · report
forward in 75 papers stllm/models/peft_model.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 515cc4dd73dc0caf · report
Repository not located 3 samples, 1 ran (1 instrument-failed) · 2 unverified
3 samples are 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.
split_to_even_chunks in 41 papers repository not located ran; the contract check could not run (RAISES) · metamorphic tier: well formed licence of this copy not recorded · 10893c4608c08075 · report fixture: signature indices:list/1/int, lengths:dict/0/?, num_chunks:int/0/int · returned list · level recorded, date not recorded
get_mm_adapter_state_maybe_zero_3 in 44 papers repository not located unverified licence of this copy not recorded · bb35e3ac741bb2c9 · report
maybe_zero_3 in 74 papers repository not located unverified licence of this copy not recorded · 735025744c1ab0cf · report

Tasks

Reading ComprehensionVideo Question AnsweringVideo UnderstandingVideo-based Generative Performance Benchmarking

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Question Answering MVBench ST-LLM Avg. 54.9 #12 of 22 Archive leaderboard report
Video Question Answering TVBench ST-LLM Average Accuracy 35.7 #26 of 28 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B Consistency 2.81 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B Contextual Understanding 3.74 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B Correctness of Information 3.23 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B Detail Orientation 3.05 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B Temporal Understanding 2.93 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct ST-LLM-7B mean 3.15 #9 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking (Consistency) VideoInstruct ST-LLM gpt-score 2.81 #8 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Contextual Understanding) VideoInstruct ST-LLM gpt-score 3.74 #5 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Correctness of Information) VideoInstruct ST-LLM gpt-score 3.23 #7 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Detail Orientation)) VideoInstruct ST-LLM gpt-score 3.05 #5 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Temporal Understanding) VideoInstruct ST-LLM gpt-score 2.93 #2 of 18 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA ST-LLM Accuracy 50.9 #10 of 28 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA ST-LLM Confidence Score 3.3 #10 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA ST-LLM Accuracy 63.2 #10 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA ST-LLM Confidence Score 3.4 #10 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA ST-LLM Accuracy 74.6 #12 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA ST-LLM Confidence Score 3.9 #12 of 28 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