Lab lands 7 NeurIPS main-track acceptances spanning foundation models, RL traffic control and optimizers
flosalim · x · 2026-09-25
Researcher flosalim congratulates students, postdocs and collaborators on 7 NeurIPS main-track acceptances, including Gluco FM (with Google), programmatic policies distilled from RL-based traffic signal controllers, the MAGE multi-agent self-evolution framework with co-evolutionary knowledge graphs, robust residual correction for time series forecasting, the FOGO forgetting-aware orthogonalization optimizer, AnyMo geometry-aware human motion modeling, and OOD detection in continual learning. D&B track results are still pending.
More from Research
- First Recorded LLM NetHack Ascension: GPT 6 Astra Wins in 37,140 Turns — egrefen · 2026-09-25
- GPT-Image-Edit-1.5M Dataset Accepted at NeurIPS Dataset Track After Multiple Rejections — cihangxie · 2026-09-25
- Researcher Praises TerminalBench Verifiers, Says Eval Design Has Grown Far More Involved in a Year — AnkaReuel · 2026-09-25
- Argon Robotics Trains Robot Policy 4x Faster Than Teleop Data, 95%+ Success Over 500 Real Runs — chris_j_paxton · 2026-09-25
- AVO agents evolve attention kernels beating FlashAttention-4 by 10.5% on B200 — bingxu_ · 2026-09-25
- One Neuron Is Enough to Bypass LLM Safety Alignment, NeurIPS 2026 Paper Shows — jonasgeiping · 2026-09-25