1 | Gabriele Sarti

1

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

We evaluate the rebus-solving capabilities of large language models on a new Italian dataset.

Multi-property Steering of Large Language Models with Dynamic Activation Composition

We propose Dynamic Activation Composition, an adaptive approach for multi-property activation steering of LLMs

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

MIRAGE uses model internals for faithful answer attribution in retrieval-augmented generation applications.

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

IT5s are the first encoder-decoder transformers pretrained on more than 40 billion Italian words.

DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers

We propose DecoderLens, a method to interpret the iterative refinement of representations in encoder-decoder Transformer models.

Quantifying the Plausibility of Context Reliance in Neural Machine Translation

We introduce PECoRe, an interpretability framework for identifying context dependence in language model generations.

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

We introduce Retrieval and Attribute-Marking enhanced Prompting (RAMP) to perform attribute-controlled MT with multilingual LLMs.

Inseq: An Interpretability Toolkit for Sequence Generation Models

We present Inseq, a Python library to democratize access to interpretability analyses of sequence generation models.

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

DivEMT is a publicly available post-editing study of Neural Machine Translation over a typologically diverse set of target languages.

Contrastive Language-Image Pre-training for the Italian Language

We present the first CLIP model for the Italian Language (CLIP-Italian), trained on more than 1.4 million image-text pairs.