Covariance-Aware Newton-Schulz (CovNS), which estimates gradient directional stability via an exponential moving average of the gradient outer product and attenuates volatile directions before orthogonalization and outperforms Muon and all baselines across all retrieval metrics on both datasets with minimal overhead.
Jin-Xin Hu, Hao Deng, Hai-Bo Xing et al.· Proceedings of the 20th ACM...· 0 citations
Autoregressive sequential recommendation trains transformer models on tokenized user behavior via cross-entropy. The Muon optimizer projects gradient matrices onto the Stiefel manifold through Newton-Schulz (NS) orthogonalization, ensuring uniform utilization of all parameter directions. However, we identify that this...
Jin-Xin Hu, Hao Deng, Hai-Bo Xing et al.· Proceedings of the 20th ACM...· 0 citations
This work proposes Astar, a training-based approach that learns a specialized evolution-guiding model from the abundant iteration histories of industrial systems, and proposes a pipeline that turns noisy historical commits into a large, clean evolutionary corpus via pairwise sample expansion and noise filtering.
Jin-Xin Hu, Hao Deng, Hai-Bo Xing et al.· 2 citations
A series of models are explored that connect item Semantic IDs with natural language in a unified representation space and seek to unlock the potential of natural-language chain-of-thought (CoT) reasoning for recommendation, including OneReason.
CMRec is proposed, a cross-country GR framework that injects cross-country supervision at the data level via dual-constrained, context-aware code-mixing that substantially improves recommendation quality in data-sparse countries while preserving performance in data-rich countries.
Yuan Gao, Hao Deng, Hai-Bo Xing et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.