# Predicting Sentence Acceptability Judgments in Multimodal Contexts

arXiv: https://arxiv.org/abs/2602.20918 · HTML: https://mlsanity.com/paper/2602.20918/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 15 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 7 of 7)

- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [Language models align with human judgments on key grammatical constructions](https://mlsanity.com/paper/2402.01676/) (2024)
- [How Furiously Can Colourless Green Ideas Sleep? Sentence Acceptability in Context](https://mlsanity.com/paper/2004.00881/) (2020)
- [Log Probabilities Are a Reliable Estimate of Semantic Plausibility in Base and Instruction-Tuned Language Models](https://mlsanity.com/paper/2403.14859/) (2024)
- [How to Make the Most of LLMs' Grammatical Knowledge for Acceptability Judgments](https://mlsanity.com/paper/2408.09639/) (2024)

## Built on by (most cited first, 1 of 1)

- [LLM and Human Modes of Representation](https://mlsanity.com/paper/2606.21616/) (2026)

## Abstract (arXiv)

Previous work has examined the capacity of deep neural networks (DNNs), particularly transformers, to predict human sentence acceptability judgments, both independently of context, and in document contexts. We consider the effect of prior exposure to visual images (i.e., visual context) on these judgments for humans and large language models (LLMs). Our results suggest that, in contrast to textual context, visual images appear to have little if any impact on human acceptability ratings. However, LLMs display the compression effect seen in previous work on human judgments in document contexts. Different sorts of LLMs are able to predict human acceptability judgments to a high degree of accuracy, but in general, their performance is slightly better when visual contexts are removed. Moreover, the distribution of LLM judgments varies among models, with Qwen resembling human patterns, and others diverging from them. LLM-generated predictions on sentence acceptability are highly correlated with their normalised log probabilities in general. However, the correlations decrease when visual contexts are present, suggesting that a higher gap exists between the internal representations of LLMs and their generated predictions in the presence of visual contexts. Our experimental work suggests interesting points of similarity and of difference between human and LLM processing of sentences in multimodal contexts.