The face of Gabriel Recchia.I’m Gabriel Recchia, a cognitive scientist at the University of Cambridge’s Winton Centre for Risk and Evidence Communication, where I work on how to communicate information in ways that support comprehension and informed decision-making; I also lead on user testing research and evaluation of patient-friendly genetic reports and the NHS: Predict family of prognostic tools. I have spent much of my career involved in research investigating the capabilities, properties, and applications of distributional models trained on large volumes of text, and have continued this while at the Winton Centre to explore their applications in characterizing how risk is communicated and perceived.

I have recently become increasingly interested in investigating the capabilities and properties of modern language models. As these models become more competent across a wider variety of domains, there is an urgent need to improve the degree to which their outputs are aligned with the intents of their operators. Today, we already observe examples where models prompted to complete particular tasks produce outputs that are biased in unintended ways, or are extremely sensitive to details of the prompt which are unrelated to the prompt-writer’s intent. There are promising techniques for guiding models to behave in ways that are more closely aligned with human preferences, but these models remain difficult to shepherd when performing tasks that are difficult for humans to evaluate, or when outputs are unsatisfactory in ways that are difficult for the evaluators to notice. I’m especially curious about whether ways of making a language model’s “reasoning process” more intelligible–e.g., “showing one’s work“–might have a role to play in making model behaviour more controllable, interpretable, and performant.

Previously, I was at the Centre for Research in the Arts, Social Sciences and Humanities, where I worked with distributional approaches to the analysis of large corpora of historical texts, and investigated conceptual change by attending to shifting statistical associations between words over time. This position also involved the development and testing of user interfaces for the display of complex quantitative information to individuals of various backgrounds.

I received my bachelor’s degree in Symbolic Systems from Stanford University in 2007, and my doctorate is in Cognitive Science at Indiana University, with a minor in computational linguistics and with language modelling as my content specialization.┬áPublications, skills, and previous employers are listed on my CV. Google Scholar may have the most up-to-date record of my research outputs.