large language models

INTERPRETING RANKING MODELS: FROM FEATURE ATTRIBUTIONS TO MECHANISTIC COMPREHENSION

The rapid adoption of Neural Ranking Models (NRMs) in Information Retrieval (IR) underscores the urgent need for robust interpretability methods to ensure their reliability and transparency. Despite their widespread use, how NRMs model query-document relevance remains poorly understood, limiting their application in hi...

WAIT-A-SEC: A Human-Centered Design Approach to Privacy Communication for Social Media Oversharing

Social media platforms often operate on extractive security logics that prioritize frictionless data sharing, leaving users disempowered and prone to oversharing sensitive information, without fully recognizing the associated privacy implications. We present WAIT-A-SEC (Warnings About Information Transfer And Sharing:...

(How) Do Large Language Models Understand High-Level Message Sequence Charts?

Abstract Large Language Models (LLMs) are being employed widely to automate tasks across the software development life-cycle. It is, however, unclear whether these tasks are performed consistently with respect to the semantics of the artefacts being handled. This question is particularly under-researched concerning arc...

OptiMat Alloys: a FAIR, living database of multi-principal element alloys enabled by a conversational agent

The FAIR principles have transformed how computational materials data and workflows are shared, yet existing repositories can only serve pre-computed entries — their coverage is perpetually incomplete and cannot adapt to new questions on demand. We built OptiMat Alloys, a large-language-model-powered conversational age...

CodaRAG: Connecting the Dots with Associativity Inspired by Complementary Learning

Large Language Models (LLMs) struggle with knowledge-intensive tasks due to hallucinations and fragmented reasoning over dispersed information. While Retrieval-Augmented Generation (RAG) grounds generation in external sources, existing methods often treat evidence as isolated units, failing to reconstruct the logical c...

A multi-stage validation framework for trustworthy large-scale clinical information extraction using large language models

Abstract Large language models (LLMs) show promise for extracting clinically meaningful information from unstructured health records, yet real-world deployment is constrained by limited scalable, trustworthy validation approaches. Conventional evaluation relies on annotation-intensive reference standards and incomplete...

Efficient and accurate steering of large language models through attention-guided feature learning

Activation steering, or direct manipulation of internal activations to guide large language model (LLM) responses toward specific semantic concepts, is emerging as a promising avenue for both understanding how semantic concepts are stored within LLMs and advancing LLM capabilities. Yet, existing steering methods are re...

TD-interpreter: enhancing the understanding of timing diagrams with visual-language learning

Abstract We introduce TD-Interpreter, a specialised machine learning (ML) workflow that assists engineers in understanding complex timing diagrams (TDs), originating from a third party, during their design and verification process. The outcome of this workflow is a visual question-answer environment which allows engine...

Data Optimization for LLMs: A Survey

Large language models (LLMs) have attracted great attention in both academic and industrial communities in recent years. Numerous survey studies have covered various aspects of LLMs, such as learning strategies, applications, alignment, explainability, and evaluation. However, little headway has been made in reviewing...

Representative Social Choice: From Learning Theory to AI Alignment

Social choice theory is the study of preference aggregation across a population, used both in mechanism design for human agents and in the democratic alignment of language models. In this study, we propose the representative social choice framework for the modeling of democratic representation in collective decisions,...

Does the additional attraction of a dwarf galaxy depend on its history? A pre-registered test of the quenching-time relation on 51 dwarf galaxies

Does the additional attraction of a dwarf galaxy depend on its history? A pre-registered test of the quenching-time relation on 51 dwarf galaxies. Version 1.5, 16 October 2026. The Informational Vector World Model (IVW) predicts that dwarf galaxies which stopped forming stars early show a larger velocity dispersion tha...

Usability Analysis of Configurator User Interfaces with Multimodal Large Language Models