Sebastian Jaszczur

5 papers · 2021–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🐣 Hot Topic Early Bird

Venues

ICML (2) NeurIPS (2) AAAI (1)

Top co-authors

Jakub Krajewski (3) Marek Cygan (3) Michał Krutul (3) Jan Ludziejewski (3) Maciej Pióro (3) Piotr Sankowski (2) Kamil Ciebiera (2) Kamil Adamczewski (2) Tomasz Odrzygóźdź (2) Piotr Miłoś (2)

Research topics

Deep Learning (4) Artificial Intelligence (3) Machine Learning (3) Application Areas (3) Architectures (3) Core AI (3) Natural Language Processing (2) Generation (1) Learning Types (1) Models (1) Resources & Methods (1)

Keywords

mixture of expert (3) scaling law (2) language modeling (1) efficient inference (1) context window (1) sparse attention (1) dense model (1) long context (1) memory constraint (1) context utilization (1) parameter scaling (1) model size (1) large language model (1) sparse layer (1) training token (1) continuous moe (1) cross-example aggregation (1) active parameters (1) empirical validation (1) efficiency gap (1)

Papers

Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient ICML 2025 Structured Packing in LLM Training Improves Long Context Utilization AAAI 2025 Mixture of Tokens: Continuous MoE through Cross-Example Aggregation NeurIPS 2024 Scaling Laws for Fine-Grained Mixture of Experts ICML 2024 Sparse is Enough in Scaling Transformers NeurIPS 2021