Papers by Gemma Boleda

Object Naming in Language and Vision: A Survey and a New Dataset (2020.lrec-1)

Copied to clipboard

Carina Silberer, Sina Zarrieß, Gemma Boleda

Challenge:	Object naming has been studied in Psycholinguistics, but has received little attention in Computational Linguistics.
Approach:	They propose a dataset that provides 36 name annotations for each of 25K objects in images selected from VisualGenome.
Outcome:	The proposed dataset shows that people choose certain names for objects, on average.

Probing for Referential Information in Language Models (2020.acl-main)

Copied to clipboard

Ionut-Teodor Sorodoc, Kristina Gulordava, Gemma Boleda

Semantic Text Processing

Challenge:	Neural network-based language models (LMs) have been shown to learn relevant properties of language without being explicitly trained for them.
Approach:	They extend their previous work to analyze whether language models capture anaphoric relations and pronoun-antecedent relations in English.
Outcome:	The Transformer outperforms the LSTM in all analyses.

Short-Term Meaning Shift: A Distributional Exploration (N19-1)

Copied to clipboard

Marco Del Tredici, Raquel Fernández, Gemma Boleda

Challenge:	a new study examines the phenomenon of short-term meaning shift in online communities . the authors use distributional representations to explore the phenomenon .
Approach:	They propose to use distributional representations to explore short-term meaning shift in online communities.
Outcome:	The proposed model has problems distinguishing meaning shift from referential phenomena, and measures contextual variability to remedy this.

Run Like a Girl! Sport-Related Gender Bias in Language and Vision (2023.findings-acl)

Copied to clipboard

Sophia Harrison, Eleonora Gualdoni, Gemma Boleda

Visual Data in NLP

Challenge:	Existing social biases and stereotypes against certain groups are reproduced by computational models.
Approach:	They analyze gender bias in two Language and Vision datasets to find that they underrepresent women . they hypothesize that a bias affects human naming choices for people playing sports .
Outcome:	The proposed model reproduces gender bias in two Language and Vision datasets.

Putting Words in Context: LSTM Language Models and Lexical Ambiguity (P19-1)

Copied to clipboard

Laura Aina, Kristina Gulordava, Gemma Boleda

Semantic Text Processing

Challenge:	In language, a word can contribute a very different meaning depending on the context . lexical ambiguity involves both morphosyntactic and semantic aspects .
Approach:	They propose a method to probe hidden representations for lexical and contextual information about words.
Outcome:	The proposed method shows that both types of information are represented to a large extent, but there is room for improvement for contextual information.

Why do objects have many names? A study on word informativeness in language use and lexical systems (2024.emnlp-main)

Copied to clipboard

Eleonora Gualdoni, Gemma Boleda

Challenge:	lexical systems contain many different words that can be assigned to the same object . studies on language use have explored how speakers adapt their referring expressions to communicate in context without tackling in-context communication.
Approach:	They propose a simple measure of informativeness for words and lexical systems, grounded in a visual space, and analyze color naming data for English and Mandarin Chinese.
Outcome:	The proposed system allows for a soft mapping between referents and words, taking into account both in-context communication and the structure of the lexical system.

Communication breakdown: On the low mutual intelligibility between human and neural captioning (2022.emnlp-main)

Copied to clipboard

Roberto Dessì, Eleonora Gualdoni, Francesca Franzon, Gemma Boleda, Marco Baroni

Multimodality

Challenge:	0-shot performance of a neural caption-based image retriever is higher when fed captions from a human-produced caption generator . despite the fact that the caption generator does not take the set of distractor images into account, this performance is only marginally above chance level.
Approach:	They compare the 0-shot performance of a neural caption-based image retriever with captions from a human-produced captioner.
Outcome:	The proposed model performs better when given human-produced captions or neural captions . the best pre-trained model perform better when fed captions produced by an out-of-the-box model .

NeLLCom-Lex: A Neural-agent Framework to Study the Interplay between Lexical Systems and Language Use (2025.findings-emnlp)

Copied to clipboard

Yuqing Zhang, Ecesu Ürker, Tessa Verhoef, Gemma Boleda, Arianna Bisazza

Challenge:	Lexical semantic change has been investigated with observational and experimental methods, but observational methods cannot get at causal mechanisms.
Approach:	They introduce a neural-agent framework designed to simulate semantic change by first grounding agents in a real lexical system and then manipulating their communicative needs.
Outcome:	The proposed framework simulates the evolution of a lexical system within a single generation by grounding agents in a real lexicon and manipulating their communicative needs.

Humans Meet Models on Object Naming: A New Dataset and Analysis (2020.coling-main)

Copied to clipboard

Carina Silberer, Sina Zarrieß, Matthijs Westera, Gemma Boleda

Visual Data in NLP

Challenge:	Existing object naming datasets that use only images with a bounding box are noisy . a human-like model behavior is not stable across domains, a study finds .
Approach:	They use MN v2 to verify object naming datasets with dozens of valid names per object . they find that human-like model behavior is not stable across domains .
Outcome:	The proposed model confuses people and clothing objects more frequently than humans do.

What do Entity-Centric Models Learn? Insights from Entity Linking in Multi-Party Dialogue (N19-1)

Copied to clipboard

Laura Aina, Carina Silberer, Ionut-Teodor Sorodoc, Matthijs Westera, Gemma Boleda

Knowledge Representation

Challenge:	a recent study suggests that models that incorporate a bias towards learning entity representations are not effective at modeling entities.
Approach:	They propose to use two entity-centric models for a referential task . they show they outperform the state of the art and do better on lower frequency entities .
Outcome:	The proposed models outperform the state of the art on a referential task . they do better on lower frequency entities than a counterpart model not entity-centric .

LLMs as a synthesis between symbolic and distributed approaches to language (2025.findings-emnlp)

Copied to clipboard

Gemma Boleda

Challenge:	a fierce battle is being fought between symbolic and distributed approaches to language and cognition . a recent study shows that morphosyntactic knowledge is encoded in a near-discrete fashion in LLMs .
Approach:	a new position paper examines the role of distributed and distributed approaches in language learning . authors argue that deep learning models represent a synthesis between the two traditions .
Outcome:	a new position paper shows that deep learning models for language represent a synthesis between the two traditions.

How to represent a word and predict it, too: Improving tied architectures for language modelling (D18-1)

Copied to clipboard

Kristina Gulordava, Laura Aina, Gemma Boleda

Semantic Text Processing

Challenge:	Recent state-of-the-art models use word embeddings as input and output mappings instead of tied models.
Approach:	They propose to decouple hidden state from word embedding prediction . they extend their proposed modification to word2vec models .
Outcome:	The proposed architectures achieve comparable or better results compared to previous models without tying . the proposed architecture reduces parameters, enabling more compact models and faster learning.