As semiconductor nodes shrink to 7nm and below, aging-induced delay faults (e.g., NBTI, HCI) increasingly threaten SoC reliability. To optimize resilience, designers require early-stage evaluation of fault impacts; however, traditional Fault Injection (FI) assessments are either bottlenecked by the slow execution of pr...
Zhi-Teng Chao, Jia-Tong Chang, Jie Yu et al.· International Test Conferenc...· 0 citations
Grouped GEMM (General Matrix Multiplication) is the core primitive in performing each expert’s matrix multiplications on tokens routed to them in MoE (Mixture-of-Experts) training. However, existing implementations either drop tokens or zero-pad each expert’s token slice to meet GEMM tiling and alignment requirements....
Le Xu, Yu-Tao Sun, Ying-Bo Hao et al.· Proceedings of the Internati...· 0 citations
This work presents Token-Disentangled Latent Test-Time Scaling, an inference-time framework that makes latent refinement token-role-aware and outperforms strong output-space test-time scaling baselines under matched decoded-candidate budgets.
Hao-Xuan Ma, Yi-Hao Liu, Yu-Tao Sun et al.· 0 citations
StellarTTS is introduced, a novel mobile-optimized NAR TTS framework based on a sparse temporal embedding strategy, enabling granular control of phoneme duration, pronunciation, and prosody and a semantic-aware codec that facilitates efficient single-stage decoding.
Kaicheng Luo, Xue-Fei Gong, Yu-Tao Sun et al.· Automatic Speech Recognition...· 0 citations
NapMem is introduced, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context, and suggests that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.
Yue Xu, Yutao Sun, Yihao Liu et al.· 3 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
