145 citations · 600 across the 26 of their papers we have counts for
26 papers
Gemma 4 Technical Report
Gemma Team, Sherif El Abd, Vaibhav Aggarwal +320
We introduce Gemma 4, a new generation of open-weight, natively multimodal language models in the Gemma model family. Designed to advance compute efficiency and reasoning, the Gemm…
Assay2Mol: large language model-based drug design using BioAssay context
Yifan Deng, Spencer S. Ericksen, Anthony Gitter
Scientific databases aggregate vast amounts of quantitative data alongside descriptive text. In biochemistry, molecule screening assays evaluate candidate molecules' functional res…
Spark Transformer: Reactivating Sparsity in FFN and Attention
Chong You, Kan Wu, Zhipeng Jia +16
The discovery of the lazy neuron phenomenon in trained Transformers, where the vast majority of neurons in their feed-forward networks (FFN) are inactive for each token, has spurre…
Gemma 3 Technical Report
Gemma Team, Aishwarya Kamath, Johan Ferret +209
We introduce Gemma 3, a multimodal addition to the Gemma family of lightweight open models, ranging in scale from 1 to 27 billion parameters. This version introduces vision underst…
Compression Scaling Laws:Unifying Sparsity and Quantization
Elias Frantar, Utku Evci, Wonpyo Park +2
We investigate how different compression techniques -- such as weight and activation quantization, and weight sparsity -- affect the scaling behavior of large language models (LLMs…
The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws
Tian Jin, Ahmed Imtiaz Humayun, Utku Evci +4
Pruning eliminates unnecessary parameters in neural networks; it offers a promising solution to the growing computational demands of large language models (LLMs). While many focus…
Learning Fine-grained Parameter Sharing via Sparse Tensor Decomposition
Cem Üyük, Mike Lasby, Mohamed Yassin +2
Large neural networks achieve state-of-the-art performance on many tasks, yet their sheer size hinders deployment on resource-constrained devices. Among existing compression approa…
Towards Optimal Adapter Placement for Efficient Transfer Learning
Aleksandra I. Nowak, Otniel-Bogdan Mercea, Anurag Arnab +3
Parameter-efficient transfer learning (PETL) aims to adapt pre-trained models to new downstream tasks while minimizing the number of fine-tuned parameters. Adapters, a popular appr…
Progressive Gradient Flow for Robust N:M Sparsity Training in Transformers
Abhimanyu Rajeshkumar Bambhaniya, Amir Yazdanbakhsh, Suvinay Subramanian +4
N:M Structured sparsity has garnered significant interest as a result of relatively modest overhead and improved efficiency. Additionally, this form of sparsity holds considerable…
Scaling Laws for Sparsely-Connected Foundation Models
Elias Frantar, Carlos Riquelme, Neil Houlsby +2
We explore the impact of parameter sparsity on the scaling behavior of Transformers trained on massive datasets (i.e., "foundation models"), in both vision and language domains. In…
Dynamic Sparse Training with Structured Sparsity
Mike Lasby, Anna Golubeva, Utku Evci +2
Dynamic Sparse Training (DST) methods achieve state-of-the-art results in sparse neural network training, matching the generalization of dense models while enabling sparse training…
JaxPruner: A concise library for sparsity research
Joo Hyung Lee, Wonpyo Park, Nicole Mitchell +22
This paper introduces JaxPruner, an open-source JAX-based pruning and sparse training library for machine learning research. JaxPruner aims to accelerate research on sparse neural…
Scaling Vision Transformers to 22 Billion Parameters
Mostafa Dehghani, Josip Djolonga, Basil Mustafa +39
The scaling of Transformers has driven breakthrough capabilities for language models. At present, the largest large language models (LLMs) contain upwards of 100B parameters. Visio…
The Dormant Neuron Phenomenon in Deep Reinforcement Learning
Ghada Sokar, Rishabh Agarwal, Pablo Samuel Castro +1
In this work we identify the dormant neuron phenomenon in deep reinforcement learning, where an agent's network suffers from an increasing number of inactive neurons, thereby affec…
Training Recipe for N:M Structured Sparsity with Decaying Pruning Mask
Sheng-Chun Kao, Amir Yazdanbakhsh, Suvinay Subramanian +3
Sparsity has become one of the promising methods to compress and accelerate Deep Neural Networks (DNNs). Among different categories of sparsity, structured sparsity has gained more…
The State of Sparse Training in Deep Reinforcement Learning
Laura Graesser, Utku Evci, Erich Elsen +1
The use of sparse neural networks has seen rapid growth in recent years, particularly in computer vision. Their appeal stems largely from the reduced number of parameters required…
GradMax: Growing Neural Networks using Gradient Information
Utku Evci, Bart van Merriënboer, Thomas Unterthiner +2
The architecture and the parameters of neural networks are often optimized independently, which requires costly retraining of the parameters whenever the architecture is modified.…
Head2Toe: Utilizing Intermediate Representations for Better Transfer Learning
Utku Evci, Vincent Dumoulin, Hugo Larochelle +1
Transfer-learning methods aim to improve performance in a data-scarce target domain using a model pretrained on a data-rich source domain. A cost-efficient strategy, linear probing…
Comparing Transfer and Meta Learning Approaches on a Unified Few-Shot Classification Benchmark
Vincent Dumoulin, Neil Houlsby, Utku Evci +4
Meta and transfer learning are two successful families of approaches to few-shot learning. Despite highly related goals, state-of-the-art advances in each family are measured large…
Gradient Flow in Sparse Neural Networks and How Lottery Tickets Win
Utku Evci, Yani A. Ioannou, Cem Keskin +1
Sparse Neural Networks (NNs) can match the generalization of dense NNs using a fraction of the compute/storage for inference, and also have the potential to enable efficient traini…
A Practical Sparse Approximation for Real Time Recurrent Learning
Jacob Menick, Erich Elsen, Utku Evci +3
Current methods for training recurrent neural networks are based on backpropagation through time, which requires storing a complete history of network states, and prohibits updatin…
Rigging the Lottery: Making All Tickets Winners
Utku Evci, Trevor Gale, Jacob Menick +2
Many applications require sparse neural networks due to space or inference time restrictions. There is a large body of work on training dense networks to yield sparse networks for…
Natural Language Understanding with the Quora Question Pairs Dataset
Lakshay Sharma, Laura Graesser, Nikita Nangia +1
This paper explores the task Natural Language Understanding (NLU) by looking at duplicate question detection in the Quora dataset. We conducted extensive exploration of the dataset…
The Difficulty of Training Sparse Neural Networks
Utku Evci, Fabian Pedregosa, Aidan Gomez +1
We investigate the difficulties of training sparse neural networks and make new observations about optimization dynamics and the energy landscape within the sparse regime. Recent w…