Recent & Upcoming Talks | Gabriele Sarti

Recent & Upcoming Talks

2025

Interpreting Latent Features in Large Language Models

The presentation discusses interpreting latent features in large language models (LLMs). After an introduction on mechanistic …
May 22, 2025 Room 1313.0344, Harmonie Building, University of Groningen Paper Presentation at InCLoW Reading Group
PDF Slides

QE4PE: Word-level Quality Estimation for Human Post-Editing

Word-level quality estimation (QE) detects erroneous spans in machine translations, which can direct and facilitate human post-editing. …
Apr 15, 2025 Online Invited Talk at DFKI Saarbrücken
Code Slides

Interpretability for Language Models: Current Trends and Applications

In this presentation, I will provide an overview of the interpretability research landscape and describe various promising methods for …
Mar 24, 2025 Bernoulli Institute, Faculty of Science and Engineering, University of Groningen Seminar, MSc Course on Trustworthy and Explainable AI, University of Groningen
Code Project Project Slides

Interpreting Context Usage in Generative Language Models

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information …
Mar 11, 2025 IRT Saint Exupéry, Toulouse, France ANITI Seminar
Code Project Project Slides

2024

Aprire la scatola nera dei modelli del linguaggio: rischi e opportunità

Questo intervento sarà mirato a demistificare il funzionamento dei modelli del linguaggio (Large Language Models), ed evidenziare come …
Dec 9, 2024 Hangar Teatri, Trieste, Italy AI2S Talk - Tra logica e misteri dell'Intelligenza Artificiale
Slides

Non Verbis, Sed Rebus: Large Language Models are Weak Solvers of Italian Rebuses

Oral presentation at CLiC-it 2024
Dec 5, 2024 Pisa, Italy Oral Presentation, CLiC-it 2024 - Italian Conference on Computational Linguistics
Slides

Interpretability for Language Models: Current Trends and Applications

In this presentation, I will provide an overview of the interpretability research landscape and describe various promising methods for …
Nov 5, 2024 Online Seminar, PhD Course on XAI, Sapienza University of Rome
Project Project Slides

Interpreting Context Usage in Generative Language Models with Inseq, PECoRe and MIRAGE

This presentation focuses on applying post-hoc interpretability techniques to analyze how language models (LMs) use input information …
Jul 16, 2024 Ludwig Maximilian University of Munich, Bayern, Germany CIS LMU Seminar
Code Project Project Slides

Interpreting Context Usage in Generative Language Models with Inseq and PECoRe

This talk discusses the challenges and opportunities in conducting interpretability analyses of generative language models. We begin by …
May 20, 2024 Politecnico di Torino, Piedmont, Italy Politecnico di Torino Invited Talk
Code Project Project Slides

Quantifying the Plausibility of Context Reliance in Neural Machine Translation

This talk presents the PECoRe framework for quantifying the plausibility of context reliance in neural machine translation. The …
May 17, 2024 Area Science Park, Trieste, Italy Area Science Park Seminar
Code Project Slides

Quantifying the Plausibility of Context Reliance in Neural Machine Translation

This talk presents the PECoRe framework for quantifying the plausibility of context reliance in neural machine translation. The …
Apr 26, 2024 Harmonie Building, University of Groningen, The Netherlands GroNLP Reading Group
Code Project Slides

Post-hoc Interpretability for Generative Language Models: Explaining Context Usage in Transformers

This talk discusses the challenges of interpreting generative language models and presents Inseq, a toolkit for interpreting sequence …
Mar 1, 2024 Online SheffieldNLP Invited Talk
Code Project Project Slides

2023

Explaining Language Models with Inseq

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding …
Nov 2, 2023 University of Amsterdam, Amsterdam InDeep Masterclass - Explaining Foundation Models
Gabriele Sarti, Grzegorz Chrupała, Arianna Bisazza
Code Project Slides

Post-hoc Interpretability for Language Models

This talk discusses the challenges of interpreting generative language models and presents Inseq, a toolkit for interpreting sequence …
Oct 26, 2023 eScience Center, Amsterdam eScience Center SIG-NLP Seminar
Code Project Project Slides

Post-hoc Interpretability for NLG & Inseq: an Interpretability Toolkit for Sequence Generation Models

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding …
Jul 2, 2023 L'Arboç, Tarragona, Spain Tutorial at REST-CL, Universitat Pompeu Fabra
Code Project Slides

Post-hoc Interpretability for Neural Language Models

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding …
Jun 1, 2023 University of Trieste, Italy Invited Talk at COSMO Seminars, AI-Lab UniTS
Code Project Slides

Explaining Neural Language Models from Internal Representations to Model Predictions

As language models become increasingly complex and sophisticated, the processes leading to their predictions are growing increasingly …
May 31, 2023 University of Pisa, Italy Lab at AILC Lectures on Computational Linguistics 2023
Project Lab Materials

Post-hoc Interpretability for Neural Language Models

In recent years, Transformer-based language models have achieved remarkable progress in most language generation and understanding …
May 23, 2023 University of Groningen, Groningen AILo Talk at RUG Bernoulli Institute
Code Project Slides

Inseq: An Interpretability Toolkit for Sequence Generation Models

This talk introduces the Inseq toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples …
Apr 6, 2023 Sapienza University of Rome SapienzaNLP Invited Talk
Code Project Slides

Advanced XAI Techniques and Inseq: An Interpretability Toolkit for Sequence Generation Models

This talk introduces the Inseq toolkit for interpreting sequence generation models. The usage of Inseq is illustrated with examples …
Mar 23, 2023 Radboud University, Nijmegen InDeep Consortium Meeting - March 2023
Code Project Slides

Introducing Inseq: An Interpretability Toolkit for Sequence Generation Models

After motivating the usage of interpretability methods in NLP, this talk introduces the Inseq toolkit for interpreting sequence …
Mar 10, 2023 Harmonie Building, University of Groningen GroNLP Reading Group
Code Project Slides Video

2022

Towards User-centric Interpretability of Machine Translation Models

With the astounding advances of artificial intelligence in recent years, interpretability research has emerged as a fundamental effort …
Oct 20, 2022 Harmonie Building, University of Groningen CLCG Linguistics Lunch
Slides

Towards User-centric Interpretability of NLP Models

With the astounding advances of artificial intelligence in recent years, the field of interpretability research has emerged as a …
May 18, 2022 Online Tech Talk at Translated
Slides Slideshow

2021

Empowering Human Translators via Interpretable Interactive Neural Machine Translation

Discussing the potential applications of interpretability research to the field of neural machine translation.
Dec 14, 2021 Online XAI4Debugging Workshop at NeurIPS 2021
PDF Slides Video

Characterizing Linguistic Complexity in Humans and Language Models

Presenting my work on studying different metrics of linguistic complexity and how they correlate with linguistic phenomena and learned …
Nov 5, 2021 Online Invited talk at the Coding Aperitivo of the MilaNLP Group, Bocconi University, Italy
Slides

2019

Neural Language Models: the New Frontier of Natural Language Understanding

An overview of the latest advances in the field of NLP, with a focus on neural models and language understanding.
Nov 22, 2019 Room 1A,, Building H3, University of Trieste, Italy StaTalk 2019
Slides

The Literary Ordnance: When the Writer is an AI

Discussing the applications of AI and NLP in the fields of literature and digital humanities.
Oct 31, 2019 12:00 Ridottino Teatro Miela, Trieste, Italy Trieste Science+Fiction Festival
Gabriele Sarti, Felice Dell’Orletta, Cristina Fenu
Project Slides

Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

Is it possible to induce sparseness in neural networks while preserving its performances? An overview of latest advances in making …
Jul 18, 2019 Room 4B, Building H2bis, University of Trieste, Italy Statistical Machine Learning Course Presentation
Slides