Session

· Contributors · Organizations · Search Program · Flagged · My Recommendations · Happening Now

Research Manuscript

: Computing without Waste: Lean and Scalable Neural Architectures

Session Chairs

DescriptionThis session explores cutting-edge innovations that redefine efficiency and scalability in modern AI systems. From ultra-low-bit LLM compression and energy-aware serving to adaptive MoE architectures and reliable neural design. These works advance the performance boundaries of large-scale AI across models, systems, and real-world generative applications.

Event Type

Research Manuscript

TimeMonday, July 271:30pm - 3:00pm PDT

LocationMtg Room 101B

Topics

Tracks

Similar Sessions

RM Poster Session Two

Serving Intelligence at Scale: Architectural and Runtime Innovations for Large Models

RM Poster Session Four

Presentations

1:30pm - 1:42pm PDT	Greenllm: SLO-Aware Dynamic Frequency Scaling for Energy-Efficient LLM Serving Authors Qunyou Liu Darong Huang Marina Zapater David Atienza
1:42pm - 1:55pm PDT	Ternainfer: Pushing the Compression Boundary of Ternary LLMs Towards 1.58-Bit GPU Inference Authors Jie Gu Edwin Hsing-Mean Sha Longshan Xu Yuhong Song Yunfan Chi Qingfeng Zhuge
1:55pm - 2:08pm PDT	Inferweave: Efficient LLM Inference on the MT-3000 Processor Authors Xinxin Qi Jianbin Fang Peng Zhang Yonggang Che
2:08pm - 2:21pm PDT	Sparsity- and Tolerance-Aware Temporally Redundant Neural Networks Authors Kyoji Awaki Ali Alvi Yamato Saikawa Shogo Semba Yuichi Okuyama Hiroshi Saito Hirohide Demura Yuta Takahashi Sumio Morioka Yoichi Tomioka
2:21pm - 2:34pm PDT	Smoe: Elastic MoE-Based Inference with Serverless Computing via Fine-Grained Expert Scaling Authors Xiaofei Yue Ziming Zhao Jiongchi Yu Huidong Ma Zhaoxuan Li Tingting Li Jianwei Yin
2:34pm - 2:47pm PDT	Faster-MoA: Low-Latency Tree-Structured MoA Serving with Early Exit and Agent-Aware Prefill-Decode Overlap Authors Zijun Wang Yijiahao Qi Hanqiu Chen Zishen Wan Gongjin Sun Dongyang Li Shuyi Pei Cong "Callie" Hao
2:47pm - 3:00pm PDT	Galaxydit: Efficient Video Generation with Guidance Alignment and Adaptive Proxy in Diffusion Transformers Authors Zhiye Song Steve Dai Ben Keller Brucek Khailany

63dac.conference-program.com