11 papers · 1 filter
Align, Unify, Suppress, Route: A Coherentist View of Transformer Computation
Nura Aljaafari, Andre Freitas
Mechanistic interpretability has identified transformer circuits, but lacks a shared vocabulary for describing how their functions compose across tasks and architectures. We introd…
Principles of Concept Representation in Sentence Encoders
Isabelle Mohr, John Dujany, Jonathan Souquet +1
What makes a sentence encoder produce good concept representations? We approach this through the lens of representational compositionality: an encoder supports a concept family onl…
Emergence and Localisation of Semantic Role Circuits in LLMs
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Despite displaying semantic competence, large language models' internal mechanisms that ground abstract semantic structure remain insufficiently characterised. We propose a method…
TRACE: Training and Inference-Time Interpretability Analysis for Language Models
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Understanding when and how linguistic knowledge emerges during language model training remains a central challenge for interpretability. Most existing tools are post hoc, rely on s…
Bridging Compositional and Distributional Semantics: A Survey on Latent Semantic Geometry via AutoEncoder
Yingji Zhang, Danilo S. Carvalho, André Freitas
Integrating compositional and symbolic properties into current distributional semantic spaces can enhance the interpretability, controllability, compositionality, and generalisatio…
Learning to Disentangle Latent Reasoning Rules with Language VAEs: A Systematic Study
Yingji Zhang, Marco Valentino, Danilo S. Carvalho +1
Incorporating explicit reasoning rules within the latent space of language models (LMs) offers a promising pathway to enhance generalisation, interpretability, and controllability.…
TRACE for Tracking the Emergence of Semantic Representations in Transformers
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Modern transformer models exhibit phase transitions during training, distinct shifts from memorisation to abstraction, but the mechanisms underlying these transitions remain poorly…
CARMA: Enhanced Compositionality in LLMs via Advanced Regularisation and Mutual Information Alignment
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Large language models (LLMs) struggle with compositional generalisation, limiting their ability to systematically combine learned components to interpret novel inputs. While archit…
Interpreting token compositionality in LLMs: A robustness analysis
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Understanding the internal mechanisms of large language models (LLMs) is integral to enhancing their reliability, interpretability, and inference processes. We present Constituent-…
Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis
João Pedro Gandarela, Danilo S. Carvalho, André Freitas
This work presents a novel systematic methodology to analyse the capabilities and limitations of Large Language Models (LLMs) with feedback from a formal inference engine, on logic…
Bridging Linguistic Structure and Mechanistic Interpretability for Conceptual Interpretation in Language Models
Nura Aljaafari, Danilo S. Carvalho, André Freitas
Understanding how language models compose meaning from linguistic input remains a central problem in interpretability research. Mechanistic studies have attributed functional roles…