Papers

• Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models
  D. Gould*, F.R. Ward*, A.C. Woodruff†, R. Arike†, J. Hills†, A. Serrano†, I. Caspary†, J.R. Brown†, J.J. Jiao‡, P. Leask‡, T. Stone‡, R. Potham‡, I.G. Stan‡, H. Mayne‡, S. Hellsten‡, S. Biswas‡, A. Azarbal‡, W.L. Anderson‡, E. Najt‡, R. Greenblatt, J. Stastny
  Submitted to NeurIPS 2026
• How does information access affect LLM monitors’ ability to detect sabotage?
  R. Arike*, R.M. Moreno*, R. Subramani*, S. Biswas, F.R. Ward
  ICML 2026
• C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning
  A. Mittal*, R. Arike
  GEM Workshop @ ACL 2026
• Continual Learning in LLM Agents: Capabilities, Risks, and Research Directions
  R. Subramani*, R. Arike*, O. Terry, A. Menon, Z. Jin, F.R. Ward, S. Herd
  TAIS 2026
• Evaluating Goal Drift in Language Model Agents
  R. Arike*, E. Donoway, H. Bartsch, M. Hobbhahn
  AIES 2025
• Interpreting Learned Feedback Patterns in Large Language Models
  L. Marks*, A. Abdullah*, C. Neo, R. Arike, D.S. Krueger, P. Torr, F. Barez
  NeurIPS 2024