← authors · overview

• Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retention
  2026
• The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors
  2026
• Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
  2025
• The Impact of Off-Policy Training Data on Probe Generalisation
  2025
• How Do LLMs Persuade? Linear Probes Can Uncover Persuasion Dynamics in Multi-Turn Conversations
  2025
• ICLR: In-Context Learning of Representations
  2025
• Augmentations in Graph Contrastive Learning: Current Methodological Flaws & Towards Better Practices
  2021
• Analyzing Data-Centric Properties for Graph Contrastive Learning
  2022
• Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task
  2023
• Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
  2024
• What Makes and Breaks Safety Fine-tuning? A Mechanistic Study
  2024