Methods › Computer Vision › Vision and Language Pre-Trained Models › SimVLM

Simple Visual Language Model

SimVLM

3 papers tagged archive 2025-07-28

Introduced by ZiRui Wang et al. in SimVLM: Simple Visual Language Model Pretraining with Weak Supervision

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

SimVLM is a minimalist pretraining framework to reduce training complexity by exploiting large-scale weak supervision. It is trained end-to-end with a single prefix language modeling (PrefixLM) objective. PrefixLM enables bidirectional attention within the prefix sequence, and thus it is applicable for both decoder-only and encoder-decoder sequence-to-sequence language models.

Paper Source

Papers archive 2025-07-28

3 shown of 3, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

CoCa: Contrastive Captioners are Image-Text Foundation Models
4 May 2022 · 6 repositories · arXiv:2205.01917

Syntology ran 9 of 17 samples · 8 unverified
MAGMA -- Multimodal Augmentation of Generative Models through Adapter-based Finetuning
9 Dec 2021 · 1 repository · arXiv:2112.05253

Syntology ran 10 of 15 samples · 5 unverified
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
24 Aug 2021 · 2 repositories · arXiv:2108.10904

Syntology ran 18 of 37 samples · 19 unverified · 28 pointer-only (licence)

Tasks archive 2025-07-28

17 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

Task	Papers
Image Captioning	2
Language Modeling	2
Language Modelling	2
Visual Question Answering	2
Visual Question Answering (VQA)	2
Action Classification	1
Decoder	1
Image Classification	1
In-Context Learning	1
Question Answering	1
Representation Learning	1
Retrieval	1
Video Retrieval	1
Visual Entailment	1
Visual Reasoning	1
Zero-Shot Cross-Modal Retrieval	1
Zero-Shot Transfer Image Classification	1

Usage over time archive 2025-07-28

Papers per year the archive tags with this method, by the paper's archive date (3 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Vision and Language Pre-Trained Models

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