Apertus 1.5: From Open Text Models to Open Multimodal Reasoners
Paper • 2610.15542 • Published • 26
Martin Jaggi
mjaggi
AI & ML interests
None yet
Recent Activity
authored a paper about 17 hours ago
Apertus 1.5: From Open Text Models to Open Multimodal Reasoners authored a paper about 17 hours ago
Adapting English Quality Classifiers for Multilingual LLM Pretraining Data Selection authored a paper about 17 hours ago
Toward Cross-Lingual Quality Classifiers for Multilingual Pretraining Data SelectionOrganizations
authored 20
papers about 17 hours ago
Adapting English Quality Classifiers for Multilingual LLM Pretraining Data Selection
Paper • 2610.11585 • Published
Toward Cross-Lingual Quality Classifiers for Multilingual Pretraining Data Selection
Paper • 2604.20549 • Published • 1
Federated Learning for Malware Detection in IoT Devices
Paper • 2104.09994 • Published
MXSens: Sensitivity-Aware Mixed-Precision Quantization for Efficient LLM Inference
Paper • 2607.17733 • Published
DoGE: Domain Reweighting with Generalization Estimation
Paper • 2310.15393 • Published • 1
Enhancing Multilingual LLM Pretraining with Model-Based Data Selection
Paper • 2502.10361 • Published • 2
Mitigating Unintended Memorization with LoRA in Federated Learning for LLMs
Paper • 2502.05087 • Published
Multiplication-Free Transformer Training via Piecewise Affine Operations
Paper • 2305.17190 • Published • 2
Special Properties of Gradient Descent with Large Learning Rates
Paper • 2205.15142 • Published
Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations
Paper • 2405.18392 • Published • 12
Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors
Paper • 2606.25971 • Published
Tying the Loop -- Tied Expert Layers in Mixture-of-Experts Language Models
Paper • 2606.16825 • Published
Apertus LLM Family Expansion via Distillation and Quantization
Paper • 2605.29128 • Published
Beyond URLs: Metadata Diversity and Position for Efficient LLM Pretraining
Paper • 2511.21613 • Published • 2
Learning from History for Byzantine Robust Optimization
Paper • 2012.10333 • Published
TiMoE: Time-Aware Mixture of Language Experts
Paper • 2508.08827 • Published
Training Dynamics of the Cooldown Stage in Warmup-Stable-Decay Learning Rate Scheduler
Paper • 2508.01483 • Published • 3
Masking as an Efficient Alternative to Finetuning for Pretrained Language Models
Paper • 2004.12406 • Published • 1
Advances and Open Problems in Federated Learning
Paper • 1912.04977 • Published