Lune

Profiles2020-2026

Martin Jaggi

EPFL

31Papers
18Last author
66Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers31

  1. Gradient-Normalized Smoothness for Optimization with Approximate HessiansAuthor 2 of 3ICLR 2026 · 8 citations
  2. Towards Fully FP8 GEMM LLM Training at ScaleAuthor 4 of 4NeurIPS 2025 · 13 citations
  3. Effective Interplay between Sparsity and Quantization: From Theory to PracticeAuthor 7 of 11ICLR 2025
  4. Intrinsic User-Centric Interpretability through Global Mixture of ExpertsAuthor 5 of 6ICLR 2025
  5. QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMsAuthor 6 of 9NeurIPS 2024 · 723 citations
  6. Layer-wise linear mode connectivityAuthor 5 of 5ICLR 2024 · 22 citations
  7. Spectral Preconditioning for Gradient Methods on Graded Non-convex FunctionsAuthor 3 of 3ICML 2024 · 10 citations
  8. CoBo: Collaborative Learning via Bilevel OptimizationAuthor 2 of 2NeurIPS 2024 · 8 citations
  9. On Convergence of Incremental Gradient for Non-convex Smooth FunctionsAuthor 4 of 4ICML 2024 · 6 citations
  10. Random-Access Infinite Context Length for TransformersAuthor 2 of 2NeurIPS 2023 · 207 citations
  11. Second-Order Optimization with Lazy HessiansAuthor 3 of 3ICML 2023 · 31 citations
  12. MultiMoDN - Multimodal, Multi-Task, Interpretable Modular NetworksAuthor 4 of 6NeurIPS 2023 · 27 citations
  13. Fast Attention Over Long Sequences With Dynamic Sparse Flash AttentionAuthor 3 of 4NeurIPS 2023 · 26 citations
  14. Collaborative Learning via Prediction ConsensusAuthor 3 of 3NeurIPS 2023 · 11 citations
  15. SIMSUM: Document-level Text Simplification via Simultaneous SummarizationAuthor 3 of 5ACL 2023 · 11 citations
  16. Byzantine-Robust Learning on Heterogeneous Datasets via BucketingAuthor 2 of 2ICLR 2022 · 192 citations
  17. Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated LearningAuthor 3 of 3NeurIPS 2022 · 131 citations
  18. Implicit Gradient Alignment in Distributed and Federated LearningAuthor 3 of 4AAAI 2022 · 44 citations
  19. Breaking the centralized barrier for cross-device federated learningAuthor 2 of 7NeurIPS 2021 · 113 citations
  20. Equinox: Training (for Free) on a Custom Inference AcceleratorAuthor 4 of 4MICRO 2021 · 10 citations
  21. Self-Supervised Neural Topic ModelingAuthor 2 of 3EMNLP Findings 2021 · 9 citations
  22. Understanding the effects of data parallelism and sparsity on neural network trainingAuthor 4 of 4ICLR 2021 · 8 citations
  23. Lightweight Cross-Lingual Sentence Representation LearningAuthor 4 of 5ACL 2021
  24. Obtaining Better Static Word Embeddings Using Contextual Embedding ModelsAuthor 2 of 2ACL 2021
  25. Don't Use Large Mini-batches, Use Local SGDAuthor 5 of 5ICLR 2020 · 462 citations
  26. Evaluating The Search Phase of Neural Architecture SearchAuthor 3 of 5ICLR 2020 · 370 citations
  27. Model Fusion via Optimal TransportAuthor 2 of 2NeurIPS 2020 · 330 citations
  28. Decentralized Deep Learning with Arbitrary Communication CompressionAuthor 4 of 4ICLR 2020 · 263 citations
  29. Dynamic Model Pruning with FeedbackAuthor 5 of 5ICLR 2020 · 229 citations
  30. Practical Low-Rank Communication Compression in Decentralized Deep LearningAuthor 3 of 3NeurIPS 2020 · 63 citations

Showing the 30 most recent indexed papers.

Dusk over the sea between two cliffs drawn in fine vertical lines