Martin Jaggi

Computer Science · Physical Sciences

46h-index12.9kcitations275works0.92yr avg
OpenAlex↗ORCID↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Stochastic Gradient Optimization Techniques(90), Sparse and Compressive Sensing Techniques(54), Privacy-Preserving Technologies in Data(46), Topic Modeling(45), Natural Language Processing Techniques(34)

Publications275 total

Beyond URLs: Metadata Diversity and Position for Efficient LLM Pretraining
arXiv (Cornell University)·2025Open Access
Weight Decay may matter more than muP for Learning Rate Transfer in Practice
ArXiv.org·2025Open Access
Stochastic Difference-of-Convex Optimization with Momentum
arXiv (Cornell University)·2025Open Access
Stochastic Optimization with Random Search
arXiv (Cornell University)·2025Open Access
A Split-Client Approach to Second-Order Optimization
arXiv (Cornell University)·2025Open Access
Benchmarking Optimizers for Large Language Model Pretraining
arXiv (Cornell University)·2025Open Access
TiMoE: Time-Aware Mixture of Language Experts
arXiv (Cornell University)·2025Open Access
FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language
arXiv (Cornell University)·2025Open Access
Gradient-Normalized Smoothness for Optimization with Approximate Hessians
arXiv (Cornell University)·2025Open Access
Towards Fully FP8 GEMM LLM Training at Scale
arXiv (Cornell University)·2025Open Access
URLs Help, Topics Guide: Understanding Metadata Utility in LLM Training
arXiv (Cornell University)·2025Open Access
NeuralGrok: Accelerate Grokking by Neural Gradient Transformation
arXiv (Cornell University)·2025Open Access
Can Performant LLMs Be Ethical? Quantifying the Impact of Web Crawling Opt-Outs
arXiv (Cornell University)·2025Open Access
Using Machine Learning for move sequence visualization and generation in climbing
arXiv (Cornell University)·2025Open Access
Enhancing Multilingual LLM Pretraining with Model-Based Data Selection
ArXiv.org·2025Open Access
Mitigating Unintended Memorization with LoRA in Federated Learning for LLMs
arXiv (Cornell University)·2025Open Access
Leveraging the true depth of LLMs
arXiv (Cornell University)·2025Open Access
Analyzing & Reducing the Need for Learning Rate Warmup in GPT Training
arXiv (Cornell University)·2024Open Access
Improving Stochastic Cubic Newton with Momentum
arXiv (Cornell University)·2024Open Access
On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists
arXiv (Cornell University)·2024Open Access
page 1 of 14Next →

Frequent Co-authors

Sebastian U. Stich(13), Sai Praneeth Karimireddy(9), Virginia Smith(8), Michael I. Jordan(7), Martin Takáč(7), Chenxin Ma(5), Tao Lin(4), Peter Richtárik(4), Aymeric Dieuleveut(4), Thomas Hofmann(4), Aurélien Bellet(3), Chaoyang He(3), Jakub Konečný(3), Mehryar Mohri(3), Jean-Baptiste Cordonnier(3), Claudiu Musat(3), Salman Avestimehr(3), Simon Lacoste-Julien(3), Valéria De Luca(3), Lie He(3)