Picture for Andre Freitas

Andre Freitas

Department of Computer Science, University of Manchester, United Kingdom, Idiap Research Institute, Switzerland

Diffusion Twigs with Loop Guidance for Conditional Graph Generation

Add code
Oct 31, 2024
Viaarxiv icon

SylloBio-NLI: Evaluating Large Language Models on Biomedical Syllogistic Reasoning

Add code
Oct 18, 2024
Viaarxiv icon

Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series

Add code
Oct 17, 2024
Viaarxiv icon

Gem: Gaussian Mixture Model Embeddings for Numerical Feature Distributions

Add code
Oct 09, 2024
Viaarxiv icon

Consistent Autoformalization for Constructing Mathematical Libraries

Add code
Oct 05, 2024
Figure 1 for Consistent Autoformalization for Constructing Mathematical Libraries
Figure 2 for Consistent Autoformalization for Constructing Mathematical Libraries
Figure 3 for Consistent Autoformalization for Constructing Mathematical Libraries
Figure 4 for Consistent Autoformalization for Constructing Mathematical Libraries
Viaarxiv icon

An LLM-based Knowledge Synthesis and Scientific Reasoning Framework for Biomedical Discovery

Add code
Jun 26, 2024
Figure 1 for An LLM-based Knowledge Synthesis and Scientific Reasoning Framework for Biomedical Discovery
Figure 2 for An LLM-based Knowledge Synthesis and Scientific Reasoning Framework for Biomedical Discovery
Figure 3 for An LLM-based Knowledge Synthesis and Scientific Reasoning Framework for Biomedical Discovery
Figure 4 for An LLM-based Knowledge Synthesis and Scientific Reasoning Framework for Biomedical Discovery
Viaarxiv icon

Transformer Normalisation Layers and the Independence of Semantic Subspaces

Add code
Jun 25, 2024
Figure 1 for Transformer Normalisation Layers and the Independence of Semantic Subspaces
Figure 2 for Transformer Normalisation Layers and the Independence of Semantic Subspaces
Figure 3 for Transformer Normalisation Layers and the Independence of Semantic Subspaces
Figure 4 for Transformer Normalisation Layers and the Independence of Semantic Subspaces
Viaarxiv icon

Exploring the Limits of Fine-grained LLM-based Physics Inference via Premise Removal Interventions

Add code
Apr 29, 2024
Viaarxiv icon

Improving Semantic Control in Discrete Latent Spaces with Transformer Quantized Variational Autoencoders

Add code
Feb 01, 2024
Viaarxiv icon

Empowering Cross-lingual Abilities of Instruction-tuned Large Language Models by Translation-following demonstrations

Add code
Aug 27, 2023
Viaarxiv icon