Search Results for author: Gabriele Sarti

Found 18 papers, 13 papers with code

That Looks Hard: Characterizing Linguistic Complexity in Humans and Language Models

1 code implementation NAACL (CMCL) 2021 Gabriele Sarti, Dominique Brunato, Felice Dell’Orletta

We then show the effectiveness of linguistic features when explicitly leveraged by a regression model for predicting sentence complexity and compare its results with the ones obtained by a fine-tuned neural language model.

Language Modelling Sentence

Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation

no code implementations19 Jun 2024 Jirui Qi, Gabriele Sarti, Raquel Fernández, Arianna Bisazza

In this work, we present MIRAGE --Model Internals-based RAG Explanations -- a plug-and-play approach using model internals for faithful answer attribution in RAG applications.

Question Answering Retrieval

A Primer on the Inner Workings of Transformer-based Language Models

no code implementations30 Apr 2024 Javier Ferrando, Gabriele Sarti, Arianna Bisazza, Marta R. Costa-jussà

The rapid progress of research aimed at interpreting the inner workings of advanced language models has highlighted a need for contextualizing the insights gained from years of work in this area.


DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers

no code implementations5 Oct 2023 Anna Langedijk, Hosein Mohebbi, Gabriele Sarti, Willem Zuidema, Jaap Jumelet

In recent years, many interpretability methods have been proposed to help interpret the internal states of Transformer-models, at different levels of precision and complexity.

Decoder Logical Reasoning +4

Quantifying the Plausibility of Context Reliance in Neural Machine Translation

2 code implementations2 Oct 2023 Gabriele Sarti, Grzegorz Chrupała, Malvina Nissim, Arianna Bisazza

Establishing whether language models can use contextual information in a human-plausible way is important to ensure their trustworthiness in real-world settings.

Machine Translation Translation

Let the Models Respond: Interpreting Language Model Detoxification Through the Lens of Prompt Dependence

1 code implementation1 Sep 2023 Daniel Scalena, Gabriele Sarti, Malvina Nissim, Elisabetta Fersini

Due to language models' propensity to generate toxic or hateful responses, several techniques were developed to align model generations with users' preferences.

Language Modelling reinforcement-learning

RAMP: Retrieval and Attribute-Marking Enhanced Prompting for Attribute-Controlled Translation

no code implementations26 May 2023 Gabriele Sarti, Phu Mon Htut, Xing Niu, Benjamin Hsu, Anna Currey, Georgiana Dinu, Maria Nadejde

Attribute-controlled translation (ACT) is a subtask of machine translation that involves controlling stylistic or linguistic attributes (like formality and gender) of translation outputs.

Attribute Machine Translation +4

Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation

1 code implementation28 Feb 2023 Lukas Edman, Gabriele Sarti, Antonio Toral, Gertjan van Noord, Arianna Bisazza

Pretrained character-level and byte-level language models have been shown to be competitive with popular subword models across a range of Natural Language Processing (NLP) tasks.

Machine Translation NMT +1

Inseq: An Interpretability Toolkit for Sequence Generation Models

2 code implementations27 Feb 2023 Gabriele Sarti, Nils Feldhus, Ludwig Sickert, Oskar van der Wal, Malvina Nissim, Arianna Bisazza

Past work in natural language processing interpretability focused mainly on popular classification tasks while largely overlooking generation settings, partly due to a lack of dedicated tools.

Decoder Feature Importance +3

DivEMT: Neural Machine Translation Post-Editing Effort Across Typologically Diverse Languages

1 code implementation24 May 2022 Gabriele Sarti, Arianna Bisazza, Ana Guerberof Arenas, Antonio Toral

We publicly release the complete dataset including all collected behavioral data, to foster new research on the translation capabilities of NMT systems for typologically diverse languages.

Machine Translation NMT +1

IT5: Text-to-text Pretraining for Italian Language Understanding and Generation

3 code implementations7 Mar 2022 Gabriele Sarti, Malvina Nissim

We introduce IT5, the first family of encoder-decoder transformer models pretrained specifically on Italian.

Decoder Headline Generation +5

Contrastive Language-Image Pre-training for the Italian Language

1 code implementation19 Aug 2021 Federico Bianchi, Giuseppe Attanasio, Raphael Pisoni, Silvia Terragni, Gabriele Sarti, Sri Lakshmi

CLIP (Contrastive Language-Image Pre-training) is a very recent multi-modal model that jointly learns representations of images and texts.

Image Retrieval Multi-label zero-shot learning +2

A dissemination workshop for introducing young Italian students to NLP

1 code implementation NAACL (TeachingNLP) 2021 Lucio Messina, Lucia Busso, Claudia Roberta Combei, Ludovica Pannitto, Alessio Miaschi, Gabriele Sarti, Malvina Nissim

We describe and make available the game-based material developed for a laboratory run at several Italian science festivals to popularize NLP among young students.

Teaching NLP with Bracelets and Restaurant Menus: An Interactive Workshop for Italian Students

1 code implementation NAACL (TeachingNLP) 2021 Ludovica Pannitto, Lucia Busso, Claudia Roberta Combei, Lucio Messina, Alessio Miaschi, Gabriele Sarti, Malvina Nissim

To raise awareness, curiosity, and longer-term interest in young people, we have developed an interactive workshop designed to illustrate the basic principles of NLP and computational linguistics to high school Italian students aged between 13 and 18 years.

ArchiMeDe @ DANKMEMES: A New Model Architecture for Meme Detection

1 code implementation17 Dec 2020 Jinen Setpal, Gabriele Sarti

We introduce ArchiMeDe, a multimodal neural network-based architecture used to solve the DANKMEMES meme detections subtask at the 2020 EVALITA campaign.

Domain Adaptation

UmBERTo-MTSA @ AcCompl-It: Improving Complexity and Acceptability Prediction with Multi-task Learning on Self-Supervised Annotations

1 code implementation10 Nov 2020 Gabriele Sarti

This work describes a self-supervised data augmentation approach used to improve learning models' performances when only a moderate amount of labeled data is available.

Data Augmentation Multi-Task Learning

ETC-NLG: End-to-end Topic-Conditioned Natural Language Generation

1 code implementation25 Aug 2020 Ginevra Carbone, Gabriele Sarti

We first test the effectiveness of our approach in a low-resource setting for Italian, evaluating the conditioning for both topic models and gold annotations.

Attribute Computational Efficiency +2

Cannot find the paper you are looking for? You can Submit a new open access paper.