5 citations · 5 across the 4 of their papers we have counts for
7 papers
Training a Compact, Strong, and Realistic Text-to-Speech Model
Runyuan Cai, Yu Lin, Yiming Wang +1
Text-to-speech (TTS) systems are conventionally optimized for intelligibility and speaker similarity. However, neither objective captures the prosodic detail that distinguishes hum…
V-Engram: Trigger-Indexed External Memory for Modular Text-to-Image Personalization
Haoran He, Runyuan Cai, Yiming Wang +2
Pretrained text-to-image models contain broad visual knowledge, yet they cannot reliably acquire or refine a specific visual identity from only a few references while preserving co…
The Other Half of the Memory Wall: Serving 35B MoEs from SSD with Trained Routing Prediction
Yu Lin, Yiming Wang, Runyuan Cai +2
Mixture-of-experts (MoE) inference on consumer hardware is bounded by weight memory: a 35B-class model is 19.5GB at 4-bit, and sparsity shrinks the compute per token, not the bytes…
Data-Efficient On-Policy Distillation for Automatic Speech Recognition
Yu Lin, Yiming Wang, Runyuan Cai +1
Building competitive automatic speech recognition (ASR) models usually requires large-scale au- dio supervision, which makes reproduction and specialization expensive. We study Ark…
Tiny-Engram: Trigger-Indexed Concept Tables for Generative Vision
Runyuan Cai, Yiming Wang, Yu Lin +1
Current personalization methods for generative vision models typically encode new concepts through continuous adapters or weight updates, yet provide limited control over whether a…
Unifying Speech Recognition, Synthesis and Conversion with Autoregressive Transformers
Runyuan Cai, Yu Lin, Yiming Wang +2
Traditional speech systems typically rely on separate, task-specific models for text-to-speech (TTS), automatic speech recognition (ASR), and voice conversion (VC), resulting in fr…
FreqNet: A Frequency-domain Image Super-Resolution Network with Dicrete Cosine Transform
Runyuan Cai, Yue Ding, Hongtao Lu
Single image super-resolution(SISR) is an ill-posed problem that aims to obtain high-resolution (HR) output from low-resolution (LR) input, during which extra high-frequency inform…