4 papers · 1 filter
Evaluating the Relevance of Uncertainty Estimators for LLM Hallucination
Yedidia Agnimo, Anna Korba, Annabelle Blangero +2
Large language models (LLMs) are prone to hallucinations, i.e., statements unsupported by the input or training data, hindering reliable deployment. In parallel, numerous uncertain…
Adaptive Chunking: Optimizing Chunking-Method Selection for RAG
Paulo Roberto de Moura Júnior, Jean Lelong, Annabelle Blangero
The effectiveness of Retrieval-Augmented Generation (RAG) is highly dependent on how documents are chunked, that is, segmented into smaller units for indexing and retrieval. Yet, c…
In-Distribution Steering: Balancing Control and Coherence in Language Model Generation
Arthur Vogels, Benjamin Wong, Yann Choho +2
Activation steering methods control large language model (LLM) behavior by modifying internal activations at inference time. However, most existing activation steering methods rely…
Mitigating Text Toxicity with Counterfactual Generation
Milan Bhan, Jean-Noel Vittaut, Nina Achache +5
Toxicity mitigation consists in rephrasing text in order to remove offensive or harmful meaning. Neural natural language processing (NLP) models have been widely used to target and…