168 citations · 235 across the 10 of their papers we have counts for
10 papers
Slicing and Dicing: Configuring Optimal Mixtures of Experts
Margaret Li, Sneha Kudugunta, Danielle Rothermel +1
Mixture-of-Experts (MoE) architectures have become standard in large language models, yet many of their core design choices - expert count, granularity, shared experts, load balanc…
ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality
Shayne Longpre, Sneha Kudugunta, Niklas Muennighoff +6
Scaling laws research has focused overwhelmingly on English -- yet the most prominent AI models explicitly serve billions of international users. In this work, we undertake the lar…
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431
In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…
(Mis)Fitting: A Survey of Scaling Laws
Margaret Li, Sneha Kudugunta, Luke Zettlemoyer
Modern foundation models rely heavily on using scaling laws to guide crucial training decisions. Researchers often extrapolate the optimal architecture and hyper parameters setting…
Assessing biomedical knowledge robustness in large language models by query-efficient sampling attacks
R. Patrick Xian, Alex J. Lee, Satvik Lolla +4
The increasing depth of parametric domain knowledge in large language models (LLMs) is fueling their rapid deployment in real-world applications. Understanding model vulnerabilitie…
MiTTenS: A Dataset for Evaluating Gender Mistranslation
Kevin Robinson, Sneha Kudugunta, Romina Stella +2
Translation systems, including foundation models capable of translation, can produce errors that result in gender mistranslation, and such errors can be especially harmful. To meas…
MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
Sneha Kudugunta, Isaac Caswell, Biao Zhang +8
We introduce MADLAD-400, a manually audited, general domain 3T token monolingual dataset based on CommonCrawl, spanning 419 languages. We discuss the limitations revealed by self-a…
PaLM 2 Technical Report
Rohan Anil, Andrew M. Dai, Orhan Firat +125
We introduce PaLM 2, a new state-of-the-art language model that has better multilingual and reasoning capabilities and is more compute-efficient than its predecessor PaLM. PaLM 2 i…
A Loss Curvature Perspective on Training Instability in Deep Learning
Justin Gilmer, Behrooz Ghorbani, Ankush Garg +6
In this work, we study the evolution of the loss Hessian across many classification tasks in order to understand the effect the curvature of the loss has on the training dynamics.…
Beyond Distillation: Task-level Mixture-of-Experts for Efficient Inference
Sneha Kudugunta, Yanping Huang, Ankur Bapna +4
Sparse Mixture-of-Experts (MoE) has been a successful approach for scaling multilingual translation models to billions of parameters without a proportional increase in training com…