Papers › Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning

Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning

21 Jun 2024arXiv:2406.15334archive 2025-07-28

Brandon Huang, Chancharik Mitra, Assaf Arbelle, Leonid Karlinsky, Trevor Darrell, Roei Herzig

The recent success of interleaved Large Multimodal Models (LMMs) in few-shot learning suggests that in-context learning (ICL) with many examples can be promising for learning new tasks. However, this many-shot multimodal ICL setting has one crucial problem: it is fundamentally limited by the model's context length set at pretraining. The problem is especially prominent in the multimodal domain, which processes both text and images, requiring additional tokens. This motivates the need for a multimodal method to compress many shots into fewer tokens without finetuning. In this work, we enable LMMs to perform multimodal, many-shot in-context learning by leveraging Multimodal Task Vectors (MTV) -- compact implicit representations of in-context examples compressed in the model's attention heads. Specifically, we first demonstrate the existence of such MTV in LMMs and then leverage these extracted MTV to enable many-shot in-context learning for various vision-and-language tasks. Our experiments suggest that MTV can scale in performance with the number of compressed shots and generalize to similar out-of-domain tasks without additional context length for inference. Code: https://github.com/Brandon3964/MultiModal-Task-Vector

PaperPDFCodeCode 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.

Syntology holds a reference implementation of Multimodal Task Vectors, generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Multimodal Task Vectors') (how to connect); the free check have('2406.15334') needs no token.

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

Code

Syntology 19 code samples harvested from 1 repository linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 5 honoured, 2 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 19 samples from 1 repository, 16 ran (15 with no instrument failure; 1 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 1 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. 5 of the 16 samples that ran have no recorded date.

brandon3964/multimodal-task-vector officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: 19 samples, 16 ran (1 instrument-failed) · 3 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.

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

19 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 5 honoured, 2 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 3 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.

5ran · honoured contract
2ran · violated contract
1ran · our draft was wrong
8ran
3unverified

Licence: 19 of the 19 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 brandon3964/multimodal-task-vector. “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.

brandon3964/multimodal-task-vector official repository 19 samples, 16 ran (1 instrument-failed) · 3 unverified
add_extra_stop_words openai_api.py ran no licence file found · pointer only · 385a35a03fe04cde · report fixture: signature stop_words:list/1/str · returned list · level recorded 2026-09-25
box_iou in 34 papers eval_mm/evaluate_grounding.py ran · honoured contract fingerprinted no licence file found · pointer only · ce749f837424dd8c · report fixture: signature boxes1:Tensor/2/float, boxes2:Tensor/2/float · returned tuple · level recorded 2026-09-22
collate_fn in 500 papers eval_mm/evaluate_caption.py ran licence not identified · pointer only · a7ddf937ddb1ba60 · report fixture: signature inputs:list/1/dict, tokenizer:callable/0/? · returned tuple · level recorded 2026-09-24
collate_fn in 500 papers eval_mm/evaluate_grounding.py ran licence not identified · pointer only · d19fe23145498c35 · report fixture: signature batches:list/1/dict, tokenizer:callable/0/? · returned tuple · level recorded 2026-09-24
collate_fn in 500 papers eval_mm/evaluate_multiple_choice.py ran licence not identified · pointer only · 37a27553c04d476e · report fixture: signature batches:list/1/dict, pad_token_id:int/0/int · returned tuple · level recorded 2026-09-24
evaluate_exact_match_accuracy in 7 papers eval_mm/evaluate_vqa.py ran · honoured contract licence not identified · pointer only · e159a001bd4cbfaa · report fixture: signature entries:list/1/dict · returned float · level recorded, date not recorded
evaluate_relaxed_accuracy in 7 papers eval_mm/evaluate_vqa.py ran · honoured contract licence not identified · pointer only · 4ffcad52354b4806 · report fixture: signature entries:list/1/dict · returned float · level recorded, date not recorded
format_vizwiz MTV/preprocess.py ran licence not identified · pointer only · d3495f004eb282e1 · report fixture: signature all_data:list/1/str, cur_item:str/0/str, num_shot:int/0/int, model_helper:NoneType/0/?, split:str/0/str · returned tuple · level recorded 2026-09-24
levenshtein_distance in 18 papers eval_mm/infographicsvqa_eval.py ran · honoured contract fingerprinted licence not identified · pointer only · 90a8e73fa6de72a4 · report fixture: signature s1:str/0/str, s2:str/0/str · returned int · level recorded 2026-09-24
load_image in 273 papers MTV/models.py ran · violated contract licence not identified · pointer only · 3275ed06c9e635d8 · report fixture: signature image_file:BytesIO/0/? · returned Image · level recorded, date not recorded
load_images in 48 papers MTV/models.py ran · honoured contract fingerprinted licence not identified · pointer only · 8f69c6173ca1fadb · report fixture: signature image_files:list/1/BytesIO · returned list · level recorded, date not recorded
open_data MTV/preprocess.py ran licence not identified · pointer only · c62b6812691c4ecc · report fixture: signature dataset_name:str/0/str, path:str/0/str · returned list · level recorded 2026-09-24
preprocess in 284 papers
census: 11 implementations from 13 papers, 10 distinct outputs across 5 buckets, one shared input per bucket
finetune.py ran · our draft was wrong no licence file found · pointer only · 8b733c9fc590e4c1 · report fixture: signature sources:list/1/list, tokenizer:callable/0/?, max_len:int/0/int, system_message:str/0/str · returned dict · level recorded 2026-09-22
relaxed_correctness in 7 papers eval_mm/evaluate_vqa.py ran · violated contract licence not identified · pointer only · b3c8680ebd8739af · report fixture: signature target:str/0/str, prediction:str/0/str, max_relative_change:float/0/float · returned bool · level recorded, date not recorded
split_activations_by_head MTV/mtv_utils.py ran licence not identified · pointer only · d4db6c913fb49a77 · report fixture: signature activations:Tensor/3/float, model_config:dict/0/? · returned Tensor · level recorded 2026-09-24
trim_stop_words openai_api.py ran no licence file found · pointer only · 469cf12b400c2c5a · report fixture: signature response:str/0/str, stop_words:list/1/str · returned str · level recorded 2026-09-25
evaluate_method eval_mm/infographicsvqa_eval.py unverified licence not identified · pointer only · b90651bcc771b76f · report
get_format_func MTV/preprocess.py unverified licence not identified · pointer only · 9d2410fb29f7ac7e · report
load_model in 479 papers MTV/mtv_utils.py unverified licence not identified · pointer only · 2534a18f2f56fd6c · 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. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

1 assembly on this paper: 0 harvested, 1 generated.

generated Assembly for Multimodal Task Vectors (graph method: Multimodal Task Vectors), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 5 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. reinforce_gradient_estimator entry · level 3 · ran · violated contract · a4935c17c81d4d9b520cdfe474715cb670875749f9f31edad4e6faf982025d85

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="2406.15334") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Few-Shot LearningIn-Context Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionSETSoftmax

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