12 papers
A Reality Check of Inference-Time Context Augmentation Methods
Haeun Yu, Geert Heyman, Pepa Atanasova +1
Large language models (LLMs) often fail to utilise the context given in the input, resulting in hallucinations or a fallback to parametric knowledge. Context augmentation methods a…
The Reliance Hypothesis: Chain-of-Thought Faithfulness Rises as Language Resource Level Falls
Lucas Resck, Jingyi Sun, Qianli Wang +2
Chain-of-Thought (CoT) faithfulness matters for monitoring, explaining and safely deploying large language models, yet it has been studied mostly in English. Languages differ in le…
Explaining the Roles of Experts in Mixture-of-Experts LLMs for Personalization
Sekh Mainul Islam, Pepa Atanasova, Isabelle Augenstein
Personalization of Large Language Models (LLMs) is increasingly important for user-centric applications, yet prior methods mainly adapt dense LLMs, whose polysemantic neurons, each…
Claim Framing as a Driver of LLM Sycophancy
Jean Seo, Pepa Atanasova, Isabelle Augenstein
Sycophancy in Large Language Models, the tendency to align responses with user preferences over ground truth, has largely been studied using questions without a clear ground truth…
Investigating the Interplay between Contextual and Parametric Chain-of-Thought Faithfulness under Optimization
Jingyi Sun, Qianli Wang, Pepa Atanasova +2
Chain-of-Thought (CoT) faithfulness, i.e., whether CoTs genuinely reflect large language models' (LLM) underlying behavior, is typically evaluated with metrics under two disjoint p…
Multi-Step Knowledge Interaction Analysis via Rank-2 Subspace Disentanglement
Sekh Mainul Islam, Pepa Atanasova, Isabelle Augenstein
Natural Language Explanations (NLEs) describe how Large Language Models (LLMs) make decisions by drawing on external Context Knowledge (CK) and Parametric Knowledge (PK). Understan…
Evaluation Framework for Highlight Explanations of Context Utilisation in Language Models
Jingyi Sun, Pepa Atanasova, Sagnik Ray Choudhury +2
Context utilisation, the ability of Language Models (LMs) to incorporate relevant information from the provided context when generating responses, remains largely opaque to users,…
CUB: Benchmarking Context Utilisation Techniques for Language Models
Lovisa Hagström, Youna Kim, Haeun Yu +4
Incorporating external knowledge is crucial for knowledge-intensive tasks, such as question answering and fact checking. However, language models (LMs) may ignore relevant informat…
With Great Backbones Comes Great Adversarial Transferability
Erik Arakelyan, Karen Hambardzumyan, Davit Papikyan +4
Advances in self-supervised learning (SSL) for machine vision have improved representation robustness and model performance, giving rise to pre-trained backbones like \emph{ResNet}…
A Reality Check on Context Utilisation for Retrieval-Augmented Generation
Lovisa Hagström, Sara Vera Marjanović, Haeun Yu +5
Retrieval-augmented generation (RAG) helps address the limitations of parametric knowledge embedded within a language model (LM). In real world settings, retrieved information can…
Graph-Guided Textual Explanation Generation Framework
Shuzhou Yuan, Jingyi Sun, Ran Zhang +4
Natural language explanations (NLEs) are commonly used to provide plausible free-text explanations of a model's reasoning about its predictions. However, recent work has questioned…
Revealing the Parametric Knowledge of Language Models: A Unified Framework for Attribution Methods
Haeun Yu, Pepa Atanasova, Isabelle Augenstein
Language Models (LMs) acquire parametric knowledge from their training process, embedding it within their weights. The increasing scalability of LMs, however, poses significant cha…