30 citations · 69 across the 13 of their papers we have counts for
13 papers
Neural Coherence : Find higher performance to out-of-distribution tasks from few samples
Simon Guiroy, Mats Richter, Sarath Chandar +1
To create state-of-the-art models for many downstream tasks, it has become common practice to fine-tune a pre-trained large vision model. However, it remains an open question of ho…
Too Big to Fool: Resisting Deception in Language Models
Mohammad Reza Samsami, Mats Leon Richter, Juan Rodriguez +3
Large language models must balance their weight-encoded knowledge with in-context information from prompts to generate accurate responses. This paper investigates this interplay by…
BigDocs: An Open Dataset for Training Multimodal Models on Document and Code Tasks
Juan Rodriguez, Xiangru Jian, Siba Smarak Panigrahi +40
Multimodal AI has the potential to significantly enhance document-understanding tasks, such as processing receipts, understanding workflows, extracting data from documents, and sum…
CarbonSense: A Multimodal Dataset and Baseline for Carbon Flux Modelling
Matthew Fortier, Mats L. Richter, Oliver Sonnentag +1
Terrestrial carbon fluxes provide vital information about our biosphere's health and its capacity to absorb anthropogenic CO emissions. The importance of predicting carbon flux…
Simple and Scalable Strategies to Continually Pre-train Large Language Models
Adam Ibrahim, Benjamin Thérien, Kshitij Gupta +5
Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution i…
Continual Pre-Training of Large Language Models: How to (re)warm your model?
Kshitij Gupta, Benjamin Thérien, Adam Ibrahim +5
Large language models (LLMs) are routinely pre-trained on billions of tokens, only to restart the process over again once new data becomes available. A much cheaper and more effici…
Wuerstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models
Pablo Pernias, Dominic Rampas, Mats L. Richter +2
We introduce Würstchen, a novel architecture for text-to-image synthesis that combines competitive performance with unprecedented cost-effectiveness for large-scale text-to-image d…
Receptive Field Refinement for Convolutional Neural Networks Reliably Improves Predictive Performance
Mats L. Richter, Christopher Pal
Minimal changes to neural architectures (e.g. changing a single hyperparameter in a key layer), can lead to significant gains in predictive performance in Convolutional Neural Netw…
Should You Go Deeper? Optimizing Convolutional Neural Network Architectures without Training by Receptive Field Analysis
Mats L. Richter, Julius Schöning, Anna Wiedenroth +1
When optimizing convolutional neural networks (CNN) for a specific image-based task, specialists commonly overshoot the number of convolutional layers in their designs. By implicat…
Exploring the Properties and Evolution of Neural Network Eigenspaces during Training
Mats L. Richter, Leila Malihi, Anne-Kathrin Patricia Windler +1
In this work we explore the information processing inside neural networks using logistic regression probes \cite{probes} and the saturation metric \cite{featurespace_saturation}. W…
Size Matters
Mats L. Richter, Wolf Byttner, Ulf Krumnack +2
Fully convolutional neural networks can process input of arbitrary size by applying a combination of downsampling and pooling. However, we find that fully convolutional image class…
Feature Space Saturation during Training
Mats L. Richter, Justin Shenk, Wolf Byttner +2
We propose layer saturation - a simple, online-computable method for analyzing the information processing in neural networks. First, we show that a layer's output can be restricted…
Spectral Analysis of Latent Representations
Justin Shenk, Mats L. Richter, Anders Arpteg +1
We propose a metric, Layer Saturation, defined as the proportion of the number of eigenvalues needed to explain 99% of the variance of the latent representations, for analyzing the…