Learnable Weight Mixing Accelerates Training
torchcompiled · x · 2026-07-15
The author concludes that this might be his most interesting project yet:
- Representing parameters as a learnable weighted mix of SymExp and standard linear weights;
- Achieving up to a 1.42x wall-clock speedup during training;
- Ultimately being able to fuse back into standard weights for seamless deployment.
More from Research
- Datology webinar: Curation Studio matched top data baselines with under 20% compute — josh_wills · 2026-10-27
- KANLab: an interactive browser playground for Kolmogorov-Arnold Networks — Lukas Niehaus · 2026-10-27
- CMU's LLM2Cert synthesizes stability certificates verified via SMT and Lean — CarnegieMellonU · 2026-10-27
- NVIDIA open-sources FLORA, a robotics data engine ~125x cheaper than GPT-6 Astra — nvidia · 2026-10-27
- RSNA 2026 Deep Learning Labs return: eight hands-on AI classes over four days in Chicago — PRouzrokh · 2026-10-27
- Steering two orthogonal concept neurons makes a model compose "gnome" — Sauers_ · 2026-10-27