CORTEXA
← Browse
arxivcs.LG2026-06-26

Aurora: A Leverage-Aware Spectral Optimizer

Alec Dewulf, Dhruv Pai, Li Yang, Ashley Zhang, Ben Keigwin

We show that for tall matrix parameters, like projection matrices in the MLP layers, the Muon update can have row norms that are arbitrarily non-uniform. This can lead to a self-reinforcing feedback loop whereby neurons receive persistently small updates and eventually do not contribute meaningfully to network outputs. This problem is effectively mitigated by an additional row normalization step, but current methods do this in a way that moves the Muon update geometry away from the polar factor of the momentum matrix, which we find is undesirable. We propose Aurora, an optimizer that enforces row-uniformity of matrix parameter updates while respecting Muon's polar factor geometry. Aurora outperforms Muon in our pre-training experiments and, when combined with existing methods, achieves state-of-the-art performance among spectral optimizers on the optimizer track of the modded-nanoGPT speedrun. Additionally, we find that Aurora's empirical gains over Muon scale with the MLP expansion factor, suggesting that Aurora may allow for effective training of very wide MLP layers.

View free PDFSource page

Related papers

arxivcs.LGcs.SI2026-07-01

SAOT: Self-Supervised Continual Graph Learning with Structure-Aware Optimal Transport

Yuting Zhang, Yanbei Liu, Zhitao Xiao, Lei Geng, Yanwei Pang, Xiao Wang

Self-supervised Continual Graph Learning (CGL) aims to successively learn from a graph sequence with different tasks without label supervision - a paradigm that has attracted widespread attention. Most existing self-supervised CGL methods rely on instance-level consistency object…

View free PDFSource page
arxivcs.ROcs.LGeess.SY2026-07-15

Flow-aware Optimal Navigation in Unsteady Flows through Reinforcement Learning

Andrea Maria Braghin, Nicolò Botteghi, Matteo Tomasetto, Andrea Manzoni, Gabriele Cazzulani

Autonomous robotic navigation in nonstationary time-varying fluid flows remains a fundamental challenge due to partial observability and the unpredictability of realistic environments. While classical optimal control frameworks employed in robotics require unrealistic a-priori gl…

View free PDFSource page
arxivcs.LG2026-07-14

Generating Developable 3D Molecules via Pocket-Conditioned Diffusion and Property-Aware Optimization

Ruoxi Gao, Jiangweizhi Peng, Ziqi Chen, Frazier N. Baker, David C. Kombo, John L. Kane, et al.

Drug discovery and development is time-consuming and resource-intensive, motivating computational approaches such as diffusion models for de novo drug design. Many such models follow the structure-based drug design (SBDD) paradigm, generating molecules to fit a target binding poc…

View free PDFSource page
arxivcs.AIcs.CLcs.LG2026-07-19

Otap:Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories

Babak Barazandeh, Subhabrata Majumdar, George Michailidis

Large language model agents solve tasks by generating trajectories that interleave planning, tool calls, and intermediate results. Current evaluation metrics reduce such a trajectory to a binary success flag or compare it against a reference by exact matching. A success flag cann…

View free PDFSource page
arxivphysics.opticscs.LGphysics.app-ph2026-07-23

Reliability-Aware Bayesian Optimization of 1310 nm PCSELs with FDTD Verification

Jinglin Yu, Feiyang Wu, Longying Wen, Chongxian Yuan, Renjie Li, Zhaoyu Zhang

Near 1310 nm photonic-crystal surface-emitting lasers (PCSELs) are attractive narrow-beam sources for optical communication and sensing, but their final design refinement is costly. Small geometry changes simultaneously shift the band-edge resonance, cavity leakage, far-field div…

View free PDFSource page
arxivcs.LG2026-07-17

TD-DPO: Difference-Aware Preference Optimization for Mitigating Sycophancy in Clinical Autism Intervention Dialogue

Shuzhong Lai, Junhong Lai, Chenxi Li, Qing Zhou, Haifeng Li, Gang Pan, et al.

The sycophancy of large language models can increase the safety risk in intervention dialogue for autistic children. Supervised fine-tuning can somewhat reduce sycophancy, but relying solely on positive examples is often insufficient to identify and correct failure patterns. We o…

View free PDFSource page