13 papers
Voltage-to-temperature calibration of the High Altitude THz Solar telescope acquisition system
Gedeane G. S. Kenshima, Daniel R. Sousa, Tiago Giorgetti +2
The THz range has been under-explored for solar astronomy, mainly due to technological limitations. Only recently, a few new telescopes, such as the High Altitude Terahertz Solar (…
Sketched Linear Contrastive Learning: Approximation, Optimization, and Statistical Scaling
Ziyan Chen, Zhongzhu Zhou, Ding-Xuan Zhou
Scaling laws describe how learning performance varies with model size, data size, and compute. While recent theoretical work has established scaling laws for sketched linear regres…
Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation
Zhongzhu Zhou, Qingyang Wu, Junxiong Wang +4
Hybrid linear attention models offer an appealing path to faster long-context inference: they reduce the quadratic cost and KV-cache burden of full softmax attention while retainin…
OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization
Zhongzhu Zhou, Donglin Zhuang, Jisen Li +4
INT2 KV-cache quantization is attractive for long-context LLM serving, but it remains difficult to make both accurate and deployable. Simple rotations such as Hadamard transforms r…
When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
Junxiong Wang, Fengxiang Bie, Jisen Li +14
Speculative decoding can significantly accelerate LLM serving, yet most deployments today disentangle speculator training from serving, treating speculator training as a standalone…
SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving
Jinda Jia, Jisen Li, Zhongzhu Zhou +8
KV-cache memory is a major bottleneck in real-world LLM serving, where systems must simultaneously support latency-sensitive small-batch requests and high-throughput concurrent wor…