5 papers
Precision As You Need: Stochastic Computing Is a Dense Adaptive Quantizer
Haoran Jin, Kangqi Zhang, Jirong Yang +4
Matrix multiplications dominate the inference cost of modern transformer-based vision models, yet existing efficiency techniques such as post-training quantization and mixed-precis…
Fengshui: Demystifying Chiplet Ecosystem and Bespoke Neural Network Accelerator Codesign
Haoran Jin, Jirong Yang, Zhiheng Zhang +5
Modern ML workloads, with stringent latency and energy constraints, are increasingly hard to run efficiently on homogeneous commodity hardware. We argue that operator-level disaggr…
Agents' Last Exam
Yiyou Sun, Xinyang Han, Weichen Zhang +378
Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional d…
Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs
Wentao Ni, Kangqi Zhang, Zhongming Yu +7
As long-context inference becomes central to large language models (LLMs), attention over growing key-value caches emerges as a dominant decoding bottleneck, motivating sparse atte…
Mozart: A Chiplet Ecosystem-Accelerator Codesign Framework for Composable Bespoke Application Specific Integrated Circuits
Haoran Jin, Jirong Yang, Yunpeng Liu +3
Modern AI acceleration faces a fundamental challenge: conventional assumptions about memory requirements, batching effectiveness, and latency-throughput tradeoffs are systemwide ge…