Picture for Yee Whye Teh

Yee Whye Teh

University College London

Prompting Strategies for Enabling Large Language Models to Infer Causation from Correlation

Add code
Dec 18, 2024
Viaarxiv icon

Learning Loss Landscapes in Preference Optimization

Add code
Nov 10, 2024
Viaarxiv icon

Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset

Add code
Nov 06, 2024
Viaarxiv icon

L3Ms -- Lagrange Large Language Models

Add code
Oct 28, 2024
Viaarxiv icon

SymDiff: Equivariant Diffusion via Stochastic Symmetrisation

Add code
Oct 08, 2024
Viaarxiv icon

Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design

Add code
Jul 16, 2024
Figure 1 for Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design
Figure 2 for Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design
Figure 3 for Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design
Figure 4 for Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design
Viaarxiv icon

EvIL: Evolution Strategies for Generalisable Imitation Learning

Add code
Jun 15, 2024
Viaarxiv icon

RecurrentGemma: Moving Past Transformers for Efficient Open Language Models

Add code
Apr 11, 2024
Figure 1 for RecurrentGemma: Moving Past Transformers for Efficient Open Language Models
Figure 2 for RecurrentGemma: Moving Past Transformers for Efficient Open Language Models
Figure 3 for RecurrentGemma: Moving Past Transformers for Efficient Open Language Models
Figure 4 for RecurrentGemma: Moving Past Transformers for Efficient Open Language Models
Viaarxiv icon

Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts

Add code
Mar 13, 2024
Viaarxiv icon

Online Adaptation of Language Models with a Memory of Amortized Contexts

Add code
Mar 07, 2024
Viaarxiv icon