Pith. sign in

REVIEW 1 cited by

Unsupervised Vision-and-Language Pre-training Without Parallel Images and Captions

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 2010.12831 v2 pith:NMJZOMG5 submitted 2020-10-24 cs.CL cs.CVcs.LG

classification cs.CLcs.CVcs.LG
keywords pre-trainingdatamodelmodelsunsupervisedalignedamountbenchmarks
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Pre-trained contextual vision-and-language (V&L) models have achieved impressive performance on various benchmarks. However, existing models require a large amount of parallel image-caption data for pre-training. Such data are costly to collect and require cumbersome curation. Inspired by unsupervised machine translation, we investigate if a strong V&L representation model can be learned through unsupervised pre-training without image-caption corpora. In particular, we propose to conduct ``mask-and-predict'' pre-training on text-only and image-only corpora and introduce the object tags detected by an object recognition model as anchor points to bridge two modalities. We find that such a simple approach achieves performance close to a model pre-trained with aligned data, on four English V&L benchmarks. Our work challenges the widely held notion that aligned data is necessary for V&L pre-training, while significantly reducing the amount of supervision needed for V&L models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Learning Relational Tabular Data without Shared Features

    cs.LG 2025-02 reject novelty 7.0 of 10

    Leal learns cross-table alignment without shared features by treating lower training loss as evidence of correct row matching, with a cluster sampler to limit the candidate set.

Pith tools