Publications by Antoine Bosselut

A Logical Fallacy-Informed Framework for Argument Generation

Boi Faltings, Robert West, Luca Mouchel, Antoine Bosselut, Debjit Paul

Despite the remarkable performance of large language models (LLMs), they still struggle with generating logically sound arguments, resulting in potential risks such as spreading misinformation. An important factor contributing to LLMs' suboptimal performan ...

Association for Computational Linguistics2025

Evaluating Morphological Compositional Generalization in Large Language Models

Antoine Bosselut, Mahammad Ismayilzada

Large language models (LLMs) have demonstrated significant progress in various natural language generation and understanding tasks. However, their linguistic generalization capabilities remain questionable, raising doubts about whether these models learn l ...

Association for Computational Linguistics2025

Exploring Defeasibility in Causal Reasoning

Boi Faltings, Antoine Bosselut, Mahammad Ismayilzada, Debjit Paul, Lazar Milikic

Defeasibility in causal reasoning implies that the causal relationship between cause and effect can be strengthened or weakened. Namely, the causal strength between cause and effect should increase or decrease with the incorporation of strengthening argume ...

2024

Enhancing Procedural Writing Through Personalized Example Retrieval: A Case Study on Cooking Recipes

Tanja Christina Käser Jacober, Paola Mejia Domenzain, Luca Mouchel, Antoine Bosselut, Thiemo Wambsganss, Seyed Parsa Neshaei, Jibril Albachir Frej, Tatjana Nazaretsky

Writing high-quality procedural texts is a challenging task for many learners. While example-based learning has shown promise as a feedback approach, a limitation arises when all learners receive the same content without considering their individual input ...

2024

Course Recommender Systems Need to Consider the Job Market

Tanja Christina Käser Jacober, Antoine Bosselut, Jibril Albachir Frej, Syrielle Mathilde Marie Montariol

Current course recommender systems primarily leverage learner-course interactions, course content, learner preferences, and supplementary course details like instructor, institution, ratings, and reviews, to make their recommendation. However, these system ...

Association for Computing Machinery, Inc2024

Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning

Boi Faltings, Robert West, Antoine Bosselut, Debjit Paul

Large language models (LLMs) have been shown to perform better when asked to reason step-by-step before answering a question. However, it is unclear to what degree the model's final answer is faithful to the stated reasoning steps. In this paper, we perfor ...

Association for Computational Linguistics (ACL)2024

Instruction-tuned LLMs with World Knowledge are More Aligned to the Human Brain

Antoine Bosselut, Martin Schrimpf, Syrielle Mathilde Marie Montariol

Instruction-tuning is a widely adopted finetuning method that enables large language models (LLMs) to generate output that more closely resembles human responses. However, no studies have shown that instruction-tuning actually teaches LLMs to process langu ...

2024

ConGeo: Robust Cross-View Geo-Localization Across Ground View Variations

Devis Tuia, Antoine Bosselut, Li Mi, Javiera Francisca Castillo Navarro, Syrielle Mathilde Marie Montariol

Cross-view geo-localization aims at localizing a ground-level query image by matching it to its corresponding geo-referenced aerial view. In real-world scenarios, the task requires accommodating diverse ground images captured by users with varying orientat ...

Springer Cham2024

ConVQG: Contrastive Visual Question Generation with Multimodal Guidance

Devis Tuia, Antoine Bosselut, Li Mi, Javiera Francisca Castillo Navarro, Syrielle Mathilde Marie Montariol

Asking questions about visual environments is a crucial way for intelligent agents to understand rich multi-faceted scenes, raising the importance of Visual Question Generation (VQG) systems. Apart from being grounded to the image, existing VQG systems can ...

Association for the Advancement of Artificial Intelligence (AAAI)2024

Training Visual Language Models with Object Detection: Grounded Change Descriptions in Satellite Images

Devis Tuia, Antoine Bosselut, Javiera Francisca Castillo Navarro, Syrielle Mathilde Marie Montariol

Recently, generalist Vision Language Models (VLMs) have shown exceptional progress in tasks previously dominated by specialized computer vision models. This becomes more prevalent when visual grounding capabilities, such as the ability to reason over input ...

Institute of Electrical and Electronics Engineers Inc.2024

Let Me Teach You: Pedagogical Foundations of Feedback for Language Models

Beatriz Maria Borges Ribeiro, Tanja Christina Käser Jacober, Antoine Bosselut

Natural Language Feedback (NLF) is an increasingly popular mechanism for aligning Large Language Models (LLMs) to human preferences. Despite the diversity of the information it can convey, NLF methods are often hand-designed and arbitrary, with little syst ...

Association for Computational Linguistics2024

“Flex Tape Can't Fix That”: Bias and Misinformation in Edited Language Models

Antoine Bosselut, Syrielle Mathilde Marie Montariol

Weight-based model editing methods update the parametric knowledge of language models post-training. However, these methods can unintentionally alter unrelated parametric knowledge representations, potentially increasing the risk of harm. In this work, we ...

Association for Computational Linguistics (ACL)2024

Discovering Knowledge-Critical Subnetworks in Pretrained Language Models

Negar Foroutan Eghlidi, Antoine Bosselut, Deniz Bayazit, Zeming Chen

Pretrained language models (LMs) encode implicit representations of knowledge in their parameters. However, localizing these representations and disentangling them from each other remains an open problem. In this work, we investigate whether pretrained lan ...

Association for Computational Linguistics2024

CRAB: Assessing the Strength of Causal Relationships Between Real-World Events

Karl Aberer, Angeliki Romanou, Antoine Bosselut, Debjit Paul, Syrielle Mathilde Marie Montariol, Léo Jules Laugier

Understanding narratives requires reasoning about the cause-and-effect relationships between events mentioned in the text. While existing foundation models yield impressive results in many NLP tasks requiring reasoning, it is unclear whether they understan ...

Association for Computational Linguistics2023

PEACOK: Persona Commonsense Knowledge for Consistent and Engaging Narratives

Beatriz Maria Borges Ribeiro, Antoine Bosselut, Deniz Bayazit, Silin Gao, Soyoung Oh

Sustaining coherent and engaging narratives requires dialogue or storytelling agents to understand how the personas of speakers or listeners ground the narrative. Specifically, these agents must infer personas of their listeners to produce statements that ...

Assoc Computational Linguistics-Acl2023

Breaking the Language Barrier: Improving Cross-Lingual Reasoning with Structured Self-Attention

Karl Aberer, Negar Foroutan Eghlidi, Mohammadreza Banaei, Antoine Bosselut

In this work, we study whether multilingual language models (MultiLMs) can transfer logical reasoning abilities to other languages when they are fine-tuned for reasoning in a different language. We evaluate the cross-lingual reasoning abilities of MultiLMs ...

Association for Computational Linguistics2023

kogito: A Commonsense Knowledge Inference Toolkit

Antoine Bosselut, Mahammad Ismayilzada

In this paper, we present kogito, an open-source tool for generating commonsense inferences about situations described in text. kogito provides an intuitive and extensible interface to interact with natural language generation models that can be used for h ...

Assoc Computational Linguistics-Acl2023

Mitigating Label Biases for In-context Learning

Antoine Bosselut, Zeming Chen

Various design settings for in-context learning (ICL), such as the choice and order of the in-context examples, can bias the model's predictions. While many studies discuss these design choices, there have been few systematic investigations into categorizi ...

Assoc Computational Linguistics-Acl2023

DISCO: Distilling Counterfactuals with Large Language Models

Antoine Bosselut, Zeming Chen, Qiyue Gao

Models trained with counterfactually augmented data learn representations of the causal structure of tasks, enabling robust generalization. However, high-quality counterfactual data is scarce for most tasks and not easily generated at scale. When crowdsour ...

Assoc Computational Linguistics-Acl2023

REFINER: Reasoning Feedback on Intermediate Representations

Boi Faltings, Maxime Jean Julien Peyrard, Robert West, Beatriz Maria Borges Ribeiro, Antoine Bosselut, Mahammad Ismayilzada, Debjit Paul

Language models (LMs) have recently shown remarkable performance on reasoning tasks by explicitly generating intermediate inferences, e.g., chain-of-thought prompting. However, these intermediate inference steps may be in- appropriate deductions from the i ...

2023