DominoBench: Evaluating Multi-Step Physical Reasoning from Still Images
The source reports DominoBench, a benchmark of 854 simulator-labelled still-image scenes in 427 minimal pairs testing whether vision-language models predict physical outcomes.
Source-based brief · Editor review not recorded · See checks and limits ↓
One primary source; other links are context, not independent corroboration.
- Primary sourceHugging Face
- Discovery linkHugging Face
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Open full article view →- Original paper+45
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
The source reports DominoBench, a benchmark of 854 simulator-labelled still-image scenes in 427 minimal pairs testing whether vision-language models predict physical outcomes. The authors tested 30 models and found that unaided models reliably predict outcomes in at most 72-75% of items for the deciding person and at most 34-40% for both twins. Accuracy rose to roughly 80-92% for frontier models when the deciding calculation and quantities were written out. Small open-weight models stayed near chance. The data, hint ladder, generator and evaluation code are described as open.
Claim explorer
Attribution, source and conditionsSource-stated points from the stored brief. Claim-level assessments have not been recorded.
Source claims not independently checked; the benchmark is limited to 2D rigid-body scenes and the reported figures differ slightly between the page summary and the excerpt.
This connects a source, an attributed statement and its qualifications. It does not establish that the claim is correct.
Questions worth asking
Open questions for further reportingWhat remains unconfirmed?
Source claims not independently checked; the benchmark is limited to 2D rigid-body scenes and the reported figures differ slightly between the page summary and the excerpt.
What material was available for this brief?
The recorded brief basis is fetched source page. These points are AINewsIQ paraphrases, not quotations or independent confirmations.
Unaided, the best models are right about the deciding person in at most about 72-75% of items.
Source: Hugging Face ↗Stored source-based brief · key point 1 · AINewsIQ paraphraseBoth twins of a minimal pair are answered correctly in at most about 34-40% of items unaided.
Source: Hugging Face ↗Stored source-based brief · key point 2 · AINewsIQ paraphraseWriting out the deciding calculation raises frontier models to roughly 82-92% on the deciding person.
Source: Hugging Face ↗Stored source-based brief · key point 3 · AINewsIQ paraphraseSmall open-weight models stay near chance even when the calculation is provided.
Source: Hugging Face ↗Stored source-based brief · key point 4 · AINewsIQ paraphraseLimits from the stored brief
Source claims not independently checked; the benchmark is limited to 2D rigid-body scenes and the reported figures differ slightly between the page summary and the excerpt.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
Brief history · 1 retained version
These are changes to AINewsIQ’s own publication. Earlier history is unavailable unless recorded here. Complete retained versions are available to the editor.
- Version 1 · published
Source-based brief stored or updated. This does not establish that the publisher changed its page.
Automated record
Source observations · 1 recent captures
A changed capture means the extracted text differed. It can reflect page furniture or extraction changes; it does not establish a substantive correction. Source captures remain private.
Retained capture; no earlier comparison available.
New engagement snapshots do not mean the source itself changed. Source observations and AINewsIQ brief revisions are separate records.
Why this may matter
The results suggest fluent scene descriptions do not show that a model can predict physical outcomes, which matters for models that plan or act in the physical world.
Limits and open questions
Source claims not independently checked; the benchmark is limited to 2D rigid-body scenes and the reported figures differ slightly between the page summary and the excerpt.
Organization-specific financial, security, and operational effects require the organization's own context and review.

Join the discussion
Have a different perspective or a question about the evidence? Discussion is free with a verified email account. Every comment is reviewed before publication.
No published comments yet.
New commenters: 2 submissions per 24 hours, at least 5 minutes apart, until 3 are approved. Discussion guidelines and moderation →