Pith. sign in

REVIEW 3 cited by

A Study of Autoregressive Decoders for Multi-Tasking in Computer Vision

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2303.17376 v1 pith:WOGWWQYP submitted 2023-03-30 cs.CV cs.AIcs.LG

classification cs.CVcs.AIcs.LG
keywords decodervisionautoregressivecomputerdecodersencodermanymulti-tasking
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

There has been a recent explosion of computer vision models which perform many tasks and are composed of an image encoder (usually a ViT) and an autoregressive decoder (usually a Transformer). However, most of this work simply presents one system and its results, leaving many questions regarding design decisions and trade-offs of such systems unanswered. In this work, we aim to provide such answers. We take a close look at autoregressive decoders for multi-task learning in multimodal computer vision, including classification, captioning, visual question answering, and optical character recognition. Through extensive systematic experiments, we study the effects of task and data mixture, training and regularization hyperparameters, conditioning type and specificity, modality combination, and more. Importantly, we compare these to well-tuned single-task baselines to highlight the cost incurred by multi-tasking. A key finding is that a small decoder learned on top of a frozen pretrained encoder works surprisingly well. We call this setup locked-image tuning with decoder (LiT-decoder). It can be seen as teaching a decoder to interact with a pretrained vision model via natural language.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Active Data Curation Effectively Distills Large-Scale Multimodal Models

    cs.CV 2024-11 conditional novelty 7.0 of 10

    Selecting training data by a reference model's loss acts as an implicit distillation, and combining it with explicit distillation yields more FLOP-efficient vision-language models that beat prior SoTA on 27 benchmarks.

  2. Visual Pre-Training on Unlabeled Images using Reinforcement Learning

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Casting image-crop consistency as temporal-difference value learning improves visual representations on unlabeled web, scene, and video data.

  3. Contextual Generative Auction with Permutation-level Externalities for Online Advertising

    cs.GT 2024-12 conditional novelty 6.0 of 10

    CGA is a generative auction that builds the slot allocation autoregressively with permutation-level externalities and learns payments by minimizing ex-post regret, approximating the optimal DSIC auction.

Pith tools