Scholar
Gabriel Recchia
Google Scholar ID: XJxGdu8AAAAJ
Modulo Research Ltd
Citations & Impact
Citations
6,080
H-index
23
i10-index
34
Publications
20
Co-authors
11
list available
Publications
Resume (English only)
Academic Achievements
- Sole-authored preprint 'Teaching autoregressive language models complex tasks by demonstration' cited by papers out of Google Brain and DeepMind and discussed on Machine Learning Street Talk.
- One of four winners of the AI Impacts essay competition on the Automation of Wisdom and Philosophy (out of 90 entries).
- Third Prize recipient in the Inverse Scaling Prize competition, which focused on identifying tasks where larger language models exhibit decreased performance.
- Co-authored 'Risk perceptions of COVID-19 around the world', referenced by U.S. News, The Telegraph, The Daily Mail, BBC Future, and 130 other outlets.
- Submitted multiple papers and preprints, such as 'Confirmation bias: A challenge for scalable oversight', 'FindTheFlaws: Annotated errors for use in scalable oversight research', etc.
- Participated in the study 'Large language models are more persuasive than incentivized human persuaders' as the analysis team lead.
- Contributed question(s) that were selected for the dataset in 'Humanity's Last Exam' and became a co-author.
- Participated in the research 'Foundational challenges in assuring alignment and safety of large language models'.
- Submitted a winning task and became a co-author in 'Inverse scaling: When bigger isn't better'.
Research Experience
- Director of Modulo Research, focusing on the evaluation and alignment of large language models.
- Contributed to Usman et al.'s monumental agenda paper, 'Foundational Challenges in Assuring Alignment and Safety of Large Language Models'.
- Participated in some of a leading frontier lab's Frontier Red Team evaluation/demo projects as part of collaborations with Hidden Variable Limited.
- Led user testing research/evaluation of patient-friendly genetic reports and the widely used prognostic tool Predict: Breast Cancer at the University of Cambridge's Winton Centre for Risk and Evidence Communication.
- Investigated capabilities, properties, and applications of distributional models trained on lots of text.
- Conducted various studies of human semantic memory and how risk is communicated, perceived, and predicted.
- Wrote an alphabet book about exoplanets (sadly uncalibrated to the reading level of any child young enough to still be interested in alphabet books).
Background
- Cognitive scientist working on the evaluation and alignment of large language models as the director of Modulo Research.
Miscellany
- Wrote an alphabet book about exoplanets (sadly uncalibrated to the reading level of any child young enough to still be interested in alphabet books).
