Attribution methods are a family of techniques for tracing the influence of inputs and model components on a model's predictions. In this lecture, I will provide an overview of attribution methods, focusing in particular on shortcomings and practical applications of input attribution techniques, and their usage to analyze context usage in language models.

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

This presentation summarizes the main contributions of my PhD thesis, advocating for a user-centric perspective on interpretability research, aiming to translate theoretical advances in model understanding in practical benefits in trustworthiness and transparency for end users of these systems.

This masterclass will feature a series of insightful presentations and a hands-on tutorial focused on explainability techniques for Large Language Models (LLMs) and other deep learning architectures. Participants will gain both insights and practical experience in interpreting and understanding the inner workings of modern AI systems. In particular, my presentation provides a general introduction to popular interpretability approaches for studying large language models. Particularly, we will focus on attributional methods to identify the influence of context on model predictions and mechanistic techniques to locate and intervene in model knowledge and behaviors.

In this presentation, I will provide an overview of the interpretability research landscape and describe various promising methods for exploring and controlling the inner mechanisms of generative language models. I will start discussing post-hoc attribution technique and their usage to identify prediction-relevant inputs, showcasing their usage within our PECoRe framework for context usage attribution, and its adaptation to produce internals-based citations in retrieval-augmented generation settings (MIRAGE). The final part will present core insight from recent mechanistic interpretability literature, focusing on the construction of replacement models to build concept attribution graphs and their practical usage for monitoring LLM behaviors.

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

Science communication panel

The presentation discusses interpreting latent features in large language models (LLMs). After an introduction on mechanistic interpretability fundamentals, including feature superposition and sparse autoencoders, I discuss recent work by the Anthropic interpretability team (Ameisen et al. 2025, Lindsey et al. 2025) for extracting circuits of interpretable features from trained LLMs. Real-world investigations of Claude mechanisms, such as multi-step reasoning and multilinguality, are also analyzed.

Word-level quality estimation (QE) detects erroneous spans in machine translations, which can direct and facilitate human post-editing. While the accuracy of word-level QE systems has been assessed extensively, their usability and downstream influence on the speed, quality and editing choices of human post-editing remain understudied. Our QE4PE study investigates the impact of word-level QE on machine translation (MT) post-editing in a realistic setting involving 42 professional post-editors across two translation directions. We compare four error-span highlight modalities, including supervised and uncertainty-based word-level QE methods, for identifying potential errors in the outputs of a state-of-the-art neural MT model. Post-editing effort and productivity are estimated by behavioral logs, while quality improvements are assessed by word- and segment-level human annotation. We find that domain, language and editors' speed are critical factors in determining highlights' effectiveness, with modest differences between human-made and automated QE highlights underlining a gap between accuracy and usability in professional workflows.

In this presentation, I will provide an overview of the interpretability research landscape and describe various promising methods for exploring and controlling the inner mechanisms of generative language models. I will focus specifically on post-hoc attribution technique and their usage to identify relevant input and model components, showcasing their usage with our Inseq open-source toolkit. A practical application of attribution techniques will be presented with the PECoRe data-driven framework for context usage attribution and its adaptation to produce internals-based citations for model answers in retrieval-augmented generation settings (MIRAGE).

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

Science communication

Questo intervento sarà mirato a demistificare il funzionamento dei modelli del linguaggio (Large Language Models), ed evidenziare come lo studio di questi sistemi come 'artefatti cognitivi' possa contribuire a una migliore comprensione dei meccanismi di ragionamento (umani e non), e dei bias nella società che ci circonda.

Oral presentation at CLiC-it 2024

In this presentation, I will provide an overview of the interpretability research landscape and describe various promising methods for exploring and controlling the inner mechanisms of generative language models. I will focus specifically on post-hoc attribution technique and their usage to identify relevant input and model components, showcasing their usage with our Inseq open-source toolkit. A practical application of attribution techniques will be presented with the PECoRe data-driven framework for context usage attribution and its adaptation to produce internals-based citations for model answers in retrieval-augmented generation settings (MIRAGE).

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information throughout the generation process. We briefly introduce Inseq, our open-source toolkit designed to simplify advanced feature attribution analyses for LMs. Then, our Plausibility Evaluation of Context Reliance (PECoRe) interpretability framework is introduced to conduct data-driven analyses of context usage in LMs. In conclusion, we showcase how PECoRe can easily be adapted to retrieval-augmented generation (RAG) settings to produce internals-based citations for model answers. Our proposed Model Internals for RAG Explanations (MIRAGE) method achieves citation quality comparable to supervised answer validators with no additional training, producing citations that are faithful to actual context usage during generation.

This talk discusses the challenges and opportunities in conducting interpretability analyses of generative language models. We begin by presenting Inseq, an open-source toolkit for advanced feature attribution analyses of language models. The usage of Inseq is illustrated through examples of state-of-the-art approaches contrastive attribution, input dependence and locating factual knowledge in intermediate model representations. Then, we introduce Plausibility Evaluation of Context Reliance (PECoRe), an end-to-end interpretability framework using model internals to detect context-dependent spans in model generations and trace their prediction back to salient tokens in the available context. The usage of PECoRe is showcased on various generative tasks, including machine translation, story generation and retrieval-augmented question answering.

This talk presents the PECoRe framework for quantifying the plausibility of context reliance in neural machine translation. The framework is applied to a case study on the impact of context on the translation of gendered pronouns and other contextual phenomena in English-to-French translation. Finally, the online demo allowing users to try PECoRe with any generative language model is presented.

This talk presents the PECoRe framework for quantifying the plausibility of context reliance in neural machine translation. The framework is applied to a case study on the impact of context on the translation of gendered pronouns and other contextual phenomena in English-to-French translation. Finally, the online demo allowing users to try PECoRe with any generative language model is presented.

This talk discusses the challenges of interpreting generative language models and presents Inseq, a toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples introducing state-of-the-art approaches for interpreting language models such as contrastive attribution. Finally, the PECoRe framework is presented as a mean to evaluate the plausibility of context usage in language models.

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding tasks. However, the internal computations of these models are hardly interpretable due to their highly nonlinear structure, hindering their usage for mission-critical applications requiring trustworthiness and transparency guarantees. This presentation will introduce interpretability methods used for tracing the predictions of language models back to their inputs and discuss how these can be used to gain insights into model biases and behaviors. Several concrete examples of language model attributions will be presented throughout the presentation using the Inseq interpretability library.

This talk discusses the challenges of interpreting generative language models and presents Inseq, a toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples introducing state-of-the-art approaches for interpreting language models such as contrastive attribution. Finally, the PECoRe framework is presented as a mean to evaluate the plausibility of context usage in language models.

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding tasks. However, the internal computations of these models are hardly interpretable due to their highly nonlinear structure, hindering their usage for mission-critical applications requiring trustworthiness and transparency guarantees. This presentation will introduce interpretability methods used for tracing the predictions of language models back to their inputs and discuss how these can be used to gain insights into model biases and behaviors. Several concrete examples of language model attributions will be presented throughout the presentation using the Inseq interpretability library.

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding tasks. However, the internal computations of these models are hardly interpretable due to their highly nonlinear structure, hindering their usage for mission-critical applications requiring trustworthiness and transparency guarantees. This presentation will introduce interpretability methods used for tracing the predictions of language models back to their inputs and discuss how these can be used to gain insights into model biases and behaviors. Several concrete examples of language model attributions will be presented throughout the presentation using the Inseq interpretability library.

As language models become increasingly complex and sophisticated, the processes leading to their predictions are growing increasingly difficult to understand. Research in NLP interpretability focuses on explaining the rationales driving model predictions and is crucial for building trust and transparency in the usage of these systems in real-world scenarios. In this laboratory, we will explore various techniques for analyzing Neural Language Models, such as feature attribution methods and diagnostic classifiers. Besides common approaches to inspect models’ internal representations, we will also introduce prompting techniques to elicit model responses and motivate their usage as alternative methods for the behavioral study of model generations.

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding tasks. However, the internal computations of these models are hardly interpretable due to their highly nonlinear structure, hindering their usage for mission-critical applications requiring trustworthiness and transparency guarantees. This presentation will introduce interpretability methods used for tracing the predictions of language models back to their inputs and discuss how these can be used to gain insights into model biases and behaviors. Throughout the presentation, several concrete examples of language model attributions will be presented using the Inseq interpretability library.

This talk introduces the Inseq toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples introducing state-of-the-art approaches for interpreting language models such as contrastive attribution, tuned lenses and causal mediation analysis.

This talk introduces the Inseq toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples introducing state-of-the-art approaches for interpreting language models such as contrastive attribution, tuned lenses and causal mediation analysis.

After motivating the usage of interpretability methods in NLP, this talk introduces the Inseq toolkit for interpreting sequence generation models. The usage of Inseq is illustrated on two case studies related to gender bias in machine translation and locating factual knowledge withing GPT-2 representations.

With the astounding advances of artificial intelligence in recent years, the field of interpretability research has emerged as a fundamental effort to ensure the development of robust AI systems aligned with human values. In this talk, two perspectives on AI interpretability will be presented alongside two case studies in natural language processing. The first study leverages behavioral data and probing tasks to study the perception and encoding of linguistic complexity in humans and language models. The second introduces a user-centric interpretability perspective for neural machine translation to improve post-editing productivity and enjoyability. The need for such application-driven approaches will be emphasized in light of current challenges in faithfully evaluating advances in this field of study.

Discussing the potential applications of interpretability research to the field of neural machine translation.

Presenting my work on studying different metrics of linguistic complexity and how they correlate with linguistic phenomena and learned representations in neural language models

An overview of the latest advances in the field of NLP, with a focus on neural models and language understanding.

Discussing the applications of AI and NLP in the fields of literature and digital humanities.

Live demo