32 citations · 62 across the 21 of their papers we have counts for
21 papers
Automated Interpretability and Feature Discovery in Language Models with Agents
Arnau Marin-Llobet, Javier Ferrando
We introduce an autonomous multiagent framework for mechanistic interpretability that automates both explaining and finding internal features in large language models. The system r…
Exploring Language-Agnosticity in Function Vectors: A Case Study in Machine Translation
Nurkhan Laiyk, Gerard I. Gállego, Javier Ferrando +1
Function vectors (FVs) are vector representations of tasks extracted from model activations during in-context learning. While prior work has shown that multilingual model represent…
Language Models Can Explain Visual Features via Steering
Javier Ferrando, Enrique Lopez-Cuena, Pablo Agustin Martin-Torres +3
Sparse Autoencoders uncover thousands of features in vision models, yet explaining these features without requiring human intervention remains an open challenge. While previous wor…
Weight space Detection of Backdoors in LoRA Adapters
David Puertolas Merenciano, Ekaterina Vasyagina, Kevin Zhu +2
LoRA adapters let users fine-tune large language models (LLMs) efficiently. However, LoRA adapters are shared through open repositories like Hugging Face Hub \citep{huggingface_hub…
Putting a Face to Forgetting: Continual Learning meets Mechanistic Interpretability
Sergi Masip, Gido M. van de Ven, Javier Ferrando +1
Catastrophic forgetting in continual learning is often measured at the performance or last-layer representation level, overlooking the underlying mechanisms. We introduce a mechani…
Real-Time Detection of Hallucinated Entities in Long-Form Generation
Oscar Obeso, Andy Arditi, Javier Ferrando +3
Large language models are now routinely used in high-stakes applications where hallucinations can cause serious harm, such as medical consultations or legal advice. Existing halluc…
Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
Javier Ferrando, Oscar Obeso, Senthooran Rajamanoharan +1
Hallucinations in large language models are a widespread problem, yet the mechanisms behind whether models will hallucinate are poorly understood, limiting our ability to solve thi…
On the Similarity of Circuits across Languages: a Case Study on the Subject-verb Agreement Task
Javier Ferrando, Marta R. Costa-jussà
Several algorithms implemented by language models have recently been successfully reversed-engineered. However, these findings have been concentrated on specific tasks and models,…
A Primer on the Inner Workings of Transformer-based Language Models
Javier Ferrando, Gabriele Sarti, Arianna Bisazza +1
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 wor…
Information Flow Routes: Automatically Interpreting Language Models at Scale
Javier Ferrando, Elena Voita
Information flows by routes inside the network via mechanisms implemented in the model. These routes can be represented as graphs where nodes correspond to token representations an…
LM Transparency Tool: Interactive Tool for Analyzing Transformer Language Models
Igor Tufanov, Karen Hambardzumyan, Javier Ferrando +1
We present the LM Transparency Tool (LM-TT), an open-source interactive toolkit for analyzing the internal workings of Transformer-based language models. Differently from previousl…
Neurons in Large Language Models: Dead, N-gram, Positional
Elena Voita, Javier Ferrando, Christoforos Nalmpantis
We analyze a family of large language models in such a lightweight manner that can be done on a single GPU. Specifically, we focus on the OPT family of models ranging from 125m to…
Automating Behavioral Testing in Machine Translation
Javier Ferrando, Matthias Sperber, Hendra Setiawan +2
Behavioral testing in NLP allows fine-grained evaluation of systems by examining their linguistic capabilities through the analysis of input-output behavior. Unfortunately, existin…
Explaining How Transformers Use Context to Build Predictions
Javier Ferrando, Gerard I. Gállego, Ioannis Tsiamas +1
Language Generation Models produce words based on the previous context. Although existing methods offer input attributions as explanations for a model's prediction, it is still unc…
Toxicity in Multilingual Machine Translation at Scale
Marta R. Costa-jussà, Eric Smith, Christophe Ropers +4
Machine Translation systems can produce different types of errors, some of which are characterized as critical or catastrophic due to the specific negative impact that they can hav…
Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
Javier Ferrando, Gerard I. Gállego, Belen Alastruey +2
In Neural Machine Translation (NMT), each token prediction is conditioned on the source sentence and the target prefix (what has been previously translated at a decoding step). How…
On the Locality of Attention in Direct Speech Translation
Belen Alastruey, Javier Ferrando, Gerard I. Gállego +1
Transformers have achieved state-of-the-art results across multiple NLP tasks. However, the self-attention mechanism complexity scales quadratically with the sequence length, creat…
Measuring the Mixing of Contextual Information in the Transformer
Javier Ferrando, Gerard I. Gállego, Marta R. Costa-jussà
The Transformer architecture aggregates input information through the self-attention mechanism, but there is no clear understanding of how this information is mixed across the enti…
Attention Weights in Transformer NMT Fail Aligning Words Between Sequences but Largely Explain Model Predictions
Javier Ferrando, Marta R. Costa-jussà
This work proposes an extensive analysis of the Transformer architecture in the Neural Machine Translation (NMT) setting. Focusing on the encoder-decoder attention mechanism, we pr…
Gender Bias in Multilingual Neural Machine Translation: The Architecture Matters
Marta R. Costa-jussà, Carlos Escolano, Christine Basta +3
Multilingual Neural Machine Translation architectures mainly differ in the amount of sharing modules and parameters among languages. In this paper, and from an algorithmic perspect…
Improving accuracy and speeding up Document Image Classification through parallel systems
Javier Ferrando, Juan Luis Dominguez, Jordi Torres +5
This paper presents a study showing the benefits of the EfficientNet models compared with heavier Convolutional Neural Networks (CNNs) in the Document Classification task, essentia…