Damien de Mijolla — accepted AI-conference papers

2025

Waste Not, Want Not; Recycled Gumbel Noise Improves Consistency in Natural Language Generation

Damien De Mijolla, Hannan Saddiq, Kim Moore

NAACL 2025long

Consistency in the output of language models is critical for their reliability and practical utility. Due to their training objective, language models learn to model the full space of possible continuations, leading to outputs that can vary significantly in style, content, and tone, even for similar…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2021

Shapley explainability on the data manifold

Christopher Frye, Damien de Mijolla, Tom Begley, Laurence Cowton, Megan Stanley, Ilya Feige

ICLR 2021poster

Explainability in AI is crucial for model development, compliance with regulation, and providing operational nuance to predictions. The Shapley framework for explainability attributes a model’s predictions to its input features in a mathematically principled and model-agnostic way. However, general…

Cited by 177 Source ↗PDF ↗Scholar ↗