“here, DALL-E mini struggles to accurately represent the implicit action for the woman, capturing the apple but not reliably capturing the eating, often sticks an apple in the man’s mouth rather than the sandwich, and in most cases misses the implication that the man and woman are supposed to be eating different things.”
Our reading, the sentence that was checked:
In the authors' experiments with the sentence 'The man is eating a sandwich and the woman an apple', DALL-E Mini struggles to represent the woman's implied action: it captures the apple but not reliably the eating, often puts an apple in the man's mouth instead of the sandwich, and in most cases misses that the two are eating different things.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Three ideas from linguistics that everyone in AI should know, paragraph 53.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-three-ideas-from-linguistics-that-179852-a1
We recorded 26 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2022-06-22. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Observation
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No horizon
- Deadline state on 2026-10-29
- No horizon (the release’s reference date)
Conditions
- on the example sentence 'The man is eating a sandwich and the woman an apple' (scope, stated by the speaker)
- in the authors' experiments with Evelina Leivada (scope, stated by the speaker)
From the speaker
- The man is eating a sandwich and the woman an apple
- not reliably capturing the eating
- often sticks an apple in the man’s mouth rather than the sandwich
- in most cases misses
- the man and woman are supposed to be eating different things
Our interpretation
- This reports the authors' contemporary experiment with the quoted ellipsis prompt, rather than all possible prompts about eating.
- Score three components separately: the woman's eating action, an apple incorrectly assigned to the man's mouth, and whether each person eats the specified different food.
- Most cases means more than half. Often and not reliably are qualitative claims of recurring errors; no numerical cutoff is supplied for them.
- The surrounding phrase presently doing is interpreted as June 2022 through publication. June 1 through June 22 is an approximate observation window; the actual experiment dates are unknown.
- The prompt and experimental setting define the observation's scope. They are not antecedents of a conditional forecast.
Counts as supported if: Independently assessed original records show recurring failures to depict the woman eating, recurring apple-to-man assignment errors, and incorrect distinct food assignments in more than half of trials.
Counts as contradicted if: The complete records show reliable depiction of the woman eating an apple and the man eating a sandwich, with apple-to-man errors rare and the claimed majority failure absent. Results that validate only some components warrant a mixed assessment.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Unresolved
The record was checked, and the evidence available by the cutoff does not settle it.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
The specific experimental account remains unverified. Searches found the original assertion and related research, but no independent record establishing the woman's missing action, the man's incorrect food, and the claimed majority failure in these Mini trials. General action-composition results cannot establish those three frequencies.
Limitations
- A replication could establish reproducibility without proving what happened in the authors' experiment.
- The three reported errors must not be collapsed into one general image-quality score.
Evidence (0 items)
No evidence item settled this claim by the cutoff.
What would settle it
The complete original image set and generation records, independently scored for all three components. A documented replication of the historical model would provide additional evidence about reproducibility.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Three ideas from linguistics that everyone in AI should know
here, DALL-E mini struggles to accurately represent the implicit action for the woman, capturing the apple but not reliably capturing the eating, often sticks an apple in the man’s mouth rather than the sandwich, and in most cases misses the implication that the man and woman are supposed to be eating different things.
- Where
- paragraph 53
- Published
- 2022-06-22
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-three-ideas-from-linguistics-that-179852-a1: Current. Outcome: Unresolved. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment marcus-three-ideas-from-linguistics-that-179852-a1.