LoL lets LLMs adapt layer order per input, switching algorithms at inference
LotfiSanae · x · 2026-11-04
The LoL (Learning Order of Layers) idea lets an LLM adapt its layer execution order to each input — even learning a different layer order for each classic algorithm (Sudoku, mazes), with the order fixed at inference time to choose which algorithm the model runs. Presented in a 4-minute video and amplified by Adaption Labs co-founder Sudip Roy and scientific lead Hugo Larochelle.
More from Research
- Yann LeCun's Singapore talk breaks down JEPA from the math to robotics applications — ylecun · 2026-11-04
- Solo researcher builds foundation model from scratch using Lie groups and differential geometry — ReportProfessional57 · 2026-11-04
- Nanjing University's SeMo3D Unifies 3D Generation, Editing, and Completion in One Model — nanjinguniv · 2026-11-04
- RoboColosseum Benchmark Finds Weak Cross-Embodiment Agreement in Robot Policies — huggingface · 2026-11-04
- SpecPilot: Berkeley and NVIDIA's Agent Loop Auto-Optimizes Speculative Decoding per Deployment — UCBerkeley · 2026-11-04
- Adobe's MoviEdit Tackles Long-Form Multi-Shot Video Editing With 87K Auto-Generated Pairs — adobe · 2026-11-04