Author name cluster

Elia Bruni

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

IJCAI Conference 2024 Conference Paper

GRASP: A Novel Benchmark for Evaluating Language GRounding and Situated Physics Understanding in Multimodal Language Models

Serwan Jassim
Mario Holubar
Annika Richter
Cornelius Wolff
Xenia Ohmer
Elia Bruni

This paper presents GRASP, a novel benchmark to evaluate the language grounding and physical understanding capabilities of video-based multimodal large language models (LLMs). This evaluation is accomplished via a two-tier approach leveraging Unity simulations. The first level tests for language grounding by assessing a model's ability to relate simple textual descriptions with visual information. The second level evaluates the model's understanding of "Intuitive Physics" principles, such as object permanence and continuity. In addition to releasing the benchmark, we use it to evaluate several state-of-the-art multimodal LLMs. Our evaluation reveals significant shortcomings in the language grounding and intuitive physics capabilities of these models. Although they exhibit at least some grounding capabilities, particularly for colors and shapes, these capabilities depend heavily on the prompting strategy. At the same time, all models perform below or at the chance level of 50% in the Intuitive Physics tests, while human subjects are on average 80% correct. These identified limitations underline the importance of using benchmarks like GRASP to monitor the progress of future models in developing these competencies.

PDF Details DOI

JAIR Journal 2020 Journal Article

Compositionality Decomposed: How do Neural Networks Generalise?

Dieuwke Hupkes
Verna Dankers
Mathijs Mul
Elia Bruni

Despite a multitude of empirical studies, little consensus exists on whether neural networks are able to generalise compositionally, a controversy that, in part, stems from a lack of agreement about what it means for a neural model to be compositional. As a response to this controversy, we present a set of tests that provide a bridge between, on the one hand, the vast amount of linguistic and philosophical theory about compositionality of language and, on the other, the successful neural models of language. We collect different interpretations of compositionality and translate them into five theoretically grounded tests for models that are formulated on a task-independent level. In particular, we provide tests to investigate (i) if models systematically recombine known parts and rules (ii) if models can extend their predictions beyond the length they have seen in the training data (iii) if models’ composition operations are local or global (iv) if models’ predictions are robust to synonym substitutions and (v) if models favour rules or exceptions during training. To demonstrate the usefulness of this evaluation paradigm, we instantiate these five tests on a highly compositional data set which we dub PCFG SET and apply the resulting tests to three popular sequence-to-sequence models: a recurrent, a convolution-based and a transformer model. We provide an in-depth analysis of the results, which uncover the strengths and weaknesses of these three architectures and point to potential areas of improvement.

PDF Details DOI

IJCAI Conference 2020 Conference Paper

Compositionality Decomposed: How do Neural Networks Generalise? (Extended Abstract)

Dieuwke Hupkes
Verna Dankers
Mathijs Mul
Elia Bruni

Despite a multitude of empirical studies, little consensus exists on whether neural networks are able to generalise compositionally. As a response to this controversy, we present a set of tests that provide a bridge between, on the one hand, the vast amount of linguistic and philosophical theory about compositionality of language and, on the other, the successful neural models of language. We collect different interpretations of compositionality and translate them into five theoretically grounded tests for models that are formulated on a task-independent level. To demonstrate the usefulness of this evaluation paradigm, we instantiate these five tests on a highly compositional data set which we dub PCFG SET, apply the resulting tests to three popular sequence-to-sequence models and provide an in-depth analysis of the results.

PDF Details DOI

AAAI Conference 2020 Conference Paper

Modelling Form-Meaning Systematicity with Linguistic and Visual Features

Arie Soeteman
Dario Gutierrez
Elia Bruni
Ekaterina Shutova

Several studies in linguistics and natural language processing (NLP) pointed out systematic correspondences between word form and meaning in language. A prominent example of such systematicity is iconicity, which occurs when the form of a word is motivated by some perceptual (e. g. visual) aspect of its referent. However, the existing data-driven approaches to form-meaning systematicity modelled word meanings relying on information extracted from textual data alone. In this paper, we investigate to what extent our visual experience explains some of the form-meaning systematicity found in language. We construct word meaning representations from linguistic as well as visual data and analyze the structure and signiﬁcance of form-meaning systematicity found in English using these models. Our ﬁndings corroborate the existence of form-meaning systematicity and show that this systematicity is concentrated in localized clusters. Furthermore, applying a multimodal approach allows us to identify new patterns of systematicity that have not been previously identiﬁed with the text-based models.

PDF Details

YNIMG Journal 2015 Journal Article

Reading visually embodied meaning from the brain: Visually grounded computational models decode visual-object mental imagery induced by written text

Andrew James Anderson
Elia Bruni
Alessandro Lopopolo
Massimo Poesio
Marco Baroni

Embodiment theory predicts that mental imagery of object words recruits neural circuits involved in object perception. The degree of visual imagery present in routine thought and how it is encoded in the brain is largely unknown. We test whether fMRI activity patterns elicited by participants reading objects' names include embodied visual-object representations, and whether we can decode the representations using novel computational image-based semantic models. We first apply the image models in conjunction with text-based semantic models to test predictions of visual-specificity of semantic representations in different brain regions. Representational similarity analysis confirms that fMRI structure within ventral-temporal and lateral-occipital regions correlates most strongly with the image models and conversely text models correlate better with posterior-parietal/lateral-temporal/inferior-frontal regions. We use an unsupervised decoding algorithm that exploits commonalities in representational similarity structure found within both image model and brain data sets to classify embodied visual representations with high accuracy (8/10) and then extend it to exploit model combinations to robustly decode different brain regions in parallel. By capturing latent visual-semantic structure our models provide a route into analyzing neural representations derived from past perceptual experience rather than stimulus-driven brain activity. Our results also verify the benefit of combining multimodal data to model human-like semantic representations.

Details DOI