Sebastian Farquhar

Identifiers

• name variant Sebastian Farquhar 0.60 · backfill

Papers (7)

1. Gram: Assessing sabotage propensities via automated alignment auditing cs.LG · 2026 · author #3
2. Realistic honeypot evaluations for scheming propensity cs.LG · 2026 · author #4
3. Latent Instruction Representation Alignment: defending against jailbreaks, backdoors and undesired knowledge in LLMs cs.LG · 2026 · author #2
4. Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation cs.CL · 2023 · author #3
5. Differentially Private Continual Learning stat.ML · 2019 · author #1
6. A Unifying Bayesian View of Continual Learning stat.ML · 2019 · author #1
7. Towards Robust Evaluations of Continual Learning stat.ML · 2018 · author #1

Mentions

• 2605.30322 #3 · arxiv_oai · confidence 0.70 Sebastian Farquhar
• 2605.29729 #4 · arxiv_oai · confidence 0.70 Sebastian Farquhar

Frequent Coauthors

• Yarin Gal 4 shared papers
• David Lindner 2 shared papers
• Victoria Krakovna 2 shared papers
• Eric Easley 1 shared papers
• Lewis Ho 1 shared papers
• Lorenz Kuhn 1 shared papers
• Rohin Shah 1 shared papers