Pith. sign in

REVIEW 2 cited by

RareAct: A video dataset of unusual interactions

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 2008.01018 v1 pith:PWDOHRCS submitted 2020-08-03 cs.CV

classification cs.CV
keywords actionsrareactvideoactioncompositionalitydatasetfew-shothowto100m
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This paper introduces a manually annotated video dataset of unusual actions, namely RareAct, including actions such as "blend phone", "cut keyboard" and "microwave shoes". RareAct aims at evaluating the zero-shot and few-shot compositionality of action recognition models for unlikely compositions of common action verbs and object nouns. It contains 122 different actions which were obtained by combining verbs and nouns rarely co-occurring together in the large-scale textual corpus from HowTo100M, but that frequently appear separately. We provide benchmarks using a state-of-the-art HowTo100M pretrained video and text model and show that zero-shot and few-shot compositionality of actions remains a challenging and unsolved task.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Can Vision Language Models Understand Mimed Actions?

    cs.CL 2025-06 conditional novelty 7.0 of 10

    Vision-language models identify real actions with context far better than they identify mimed actions performed by 3D avatars, while humans are equally accurate on both.

  2. LEGO Co-builder: Exploring Fine-Grained Vision-Language Modeling for Multimodal LEGO Assembly Assistants

    cs.AI 2025-07 conditional novelty 6.0 of 10

    LEGO-VLM benchmark shows VLMs fail at fine-grained LEGO assembly state detection, but model-generated ground truth and a missing trivial baseline weaken the claim.

Pith tools