Papers › Multimodal Autoregressive Pre-training of Large Vision Encoders

Multimodal Autoregressive Pre-training of Large Vision Encoders

21 Nov 2024CVPR 2025 1arXiv:2411.14402archive 2025-07-28

Enrico Fini, Mustafa Shukor, Xiujun Li, Philipp Dufter, Michal Klein, David Haldimann, Sai Aitharaju, Victor Guilherme Turrisi da Costa, Louis Béthune, Zhe Gan, Alexander T Toshev, Marcin Eichner, Moin Nabi, Yinfei Yang, Joshua M. Susskind, Alaaeldin El-Nouby

We introduce a novel method for pre-training of large-scale vision encoders. Building on recent advancements in autoregressive pre-training of vision models, we extend this framework to a multimodal setting, i.e., images and text. In this paper, we present AIMV2, a family of generalist vision encoders characterized by a straightforward pre-training process, scalability, and remarkable performance across a range of downstream tasks. This is achieved by pairing the vision encoder with a multimodal decoder that autoregressively generates raw image patches and text tokens. Our encoders excel not only in multimodal evaluations but also in vision benchmarks such as localization, grounding, and classification. Notably, our AIMV2-3B encoder achieves 89.5% accuracy on ImageNet-1k with a frozen trunk. Furthermore, AIMV2 consistently outperforms state-of-the-art contrastive models (e.g., CLIP, SigLIP) in multimodal image understanding across diverse settings.

PaperPDFConference PDFCodeCode 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="2411.14402")

Code

Syntology Ran 0 of 3 code samples harvested from 1 repository linked to this paper; 3 have no recorded run.

By repository: official repository: 3 samples from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

apple/ml-aim officialmentioned in papermentioned on GitHubjaxNOASSERTION 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

3 samples harvested; 0 ran; 0 honoured the contract we drafted; 3 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.

3unverified

Licence: 3 of the 3 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 apple/ml-aim. “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.

accuracy apple/ml-aim/aim-v1/aim/v1/utils.py official repository unverified licence not identified · pointer only · f2939a0126b2512e · report
init_distributed_mode apple/ml-aim/aim-v1/aim/v1/utils.py official repository unverified licence not identified · pointer only · 21e85f13c1d7ea83 · report
merge_state_dicts apple/ml-aim/aim-v1/aim/v1/utils.py official repository unverified licence not identified · pointer only · 2c8b62c579ed1ba0 · report

Tasks

DecoderImage Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet AIMv2-3B (448 res) Top 1 Accuracy 89.5% #18 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-3B Top 1 Accuracy 88.5% #43 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-1B Number of params 1200M #60 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-1B Top 1 Accuracy 88.1% #60 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-H Number of params 600M #87 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-H Top 1 Accuracy 87.5% #87 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-L Number of params 300M #134 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-L Top 1 Accuracy 86.6% #134 of 1060 Archive leaderboard report
Image Classification ImageNet AIMv2-2B Number of params 2700M #1059 of 1060 Archive leaderboard report
Image Classification iNaturalist AIMv2-3B (448 res) Top 1 Accuracy 85.9 #1 of 19 Archive leaderboard report
Image Classification iNaturalist AIMv2-3B Top 1 Accuracy 81.5 #5 of 19 Archive leaderboard report
Image Classification iNaturalist AIMv2-1B Top 1 Accuracy 79.7 #8 of 19 Archive leaderboard report
Image Classification iNaturalist AIMv2-H Top 1 Accuracy 77.9 #9 of 19 Archive leaderboard report
Image Classification iNaturalist AIMv2-L Top 1 Accuracy 76 #10 of 19 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

CLIP

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