Splice-Core: Low-Order-Bit-Splicing GEMM for Energy-Efficient AI Accelerators A November 19, 2026 arXiv paper introduces Splice-Core, a temporal-unary/binary hybrid GEMM unit for AI inference accelerators whose two processing-element variants, Splice-BP and Splice-LUT, cut geometric-mean tile energy by 77.4% and 76.4% respectively versus a binary systolic array across 26 INT4 CNN and LLM workloads in 45nm. Splice-LUT also raises geometric-mean tile throughput by 44.4% and reduces geometric-mean energy-delay product by 6.13x across the 26 configurations, while the pair cut tile energy by 77.3-88.1% on W8A8 convolution and expert-projection workloads. Post-place-and-route INT4 results show 14% and 9% smaller area for Splice-BP and Splice-LUT than the systolic array. Computer Science Hardware Architecture Submitted on 19 Nov 2026 Title:Splice-Core: Low-Order-Bit-Splicing GEMM for Highly Energy-Efficient AI Inference Accelerators View PDF https://arxiv.org/pdf/2611.25617 HTML experimental https://arxiv.org/html/2611.25617v1 Abstract:Splice-Core is a temporal-unary/binary tub hybrid general matrix multiplication GEMM unit that minimizes low-order accumulation latency, while retaining value-dependent execution for high-order activation bits. We present two novel processing element PE variants: i Splice-BP utilizing bit planes, and ii Splice-LUT, which selects a shared weight multiple in a single cycle. Compared to a binary systolic array, Splice-BP/Splice-LUT achieve 77.4%/76.4% geometric-mean tile energy reductions, respectively, across 26 INT4 convolutional neural network CNN and large language model LLM workloads in 45nm. Splice-LUT further increases geometric-mean tile throughput by 44.4% and reduces geometric-mean energy-delay product EDP by 6.13x across the 26 configurations. Across the evaluated W8A8 convolution and expert-projection workloads, Splice-BP and Splice-LUT reduce tile energy by 77.3-88.1%. Post-place-and-route INT4 yields 14%/9% smaller area for Splice-BP/Splice-LUT, compared to the systolic array. Additional Features References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .