Publications

Semantic-Level Confidence Calibration of Language Models via Temperature Scaling

Published in ICLR: QUESTION Workshop, 2025

LLMs; Uncertiainty Quantification; Semantic Calibration

Recommended citation: Lamb, T. A., Ivanova, D. R., Torr, P., & Rudner, T. G. (2025). Semantic-Level Confidence Calibration of Language Models via Temperature Scaling. In ICLR Workshop: Quantify Uncertainty and Hallucination in Foundation Models: The Next Frontier in Reliable AI. https://openreview.net/pdf?id=22UNXHzBgt

Focus on this, not that! Steering LLMs with Adaptive Feature Specification

Published in ICML, 2025

LLMs; steering; spurious correlations

Recommended citation: Lamb, Tom A., et al. "Focus On This, Not That! Steering LLMs with Adaptive Feature Specification." Forty-second International Conference on Machine Learning. https://openreview.net/pdf?id=rbI5mOUA8Z

Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models

Published in NeurIPS Track on Datasets and Benchmarks., 2024

LLMs; steering; spurious correlations

Recommended citation: Hemmat, Arshia, et al. "Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models." The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track. https://arxiv.org/abs/2411.06287

Universal In-Context Approximation By Prompting Fully Recurrent Models

Published in NeurIPS, 2024

LLMs; steering; spurious correlations

Recommended citation: Petrov, A., Lamb, T., Paren, A., Torr, P. and Bibi, A., 2024. Universal in-context approximation by prompting fully recurrent models. Advances in Neural Information Processing Systems, 37, pp.72061-72093. https://proceedings.neurips.cc/paper_files/paper/2024/file/848784373188ddf641079524e89e0ac9-Paper-Conference.pdf

Faithful Knowledge Distillation [Preprint]

Published in Arxiv, 2023

Knowledge distillation; verification; robustness

Recommended citation: Lamb, Tom A., et al. "Faithful Knowledge Distillation." arXiv preprint arXiv:2306.04431 (2023). https://arxiv.org/abs/2306.04431

Tom Lamb

Publications

Semantic-Level Confidence Calibration of Language Models via Temperature Scaling

Focus on this, not that! Steering LLMs with Adaptive Feature Specification

Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models

Universal In-Context Approximation By Prompting Fully Recurrent Models

Faithful Knowledge Distillation [Preprint]