Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2604.15623 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910138522140672 |
|---|---|
| author | Wang, Weilun Wang, Zirui Li, Wantong |
| author_facet | Wang, Weilun Wang, Zirui Li, Wantong |
| contents | Neuro-symbolic AI is gaining traction in domains such as large language models, scientific discovery, and autonomous systems due to its ability to combine perception with structured reasoning. However, its deployment is often constrained by high memory demands, diverse computation patterns, and complex hardware requirements. Existing hardware platforms struggle with large on-chip memory overheads, frequent pipeline stalls, limited I/O bandwidth, and inefficient handling of nonlinear operations. To address these key computational bottlenecks, we propose Overmind, a unified neuro-symbolic architecture with cross-layer optimizations. Overmind tackles these core bottlenecks through Padé approximations for universal nonlinear functions, preemptive memory bypass that eliminates costly on-chip caches, and a complete software stack that optimizes model deployment. By reconfiguring the Padé orders for approximating nonlinear functions, we also demonstrate adaptive accuracy-performance scaling. Overmind achieves an energy efficiency of 8.1 TOPS/W and a throughput of 410 GOPS for mixed neuro-symbolic workloads with minimal model accuracy loss. Compared to existing solutions, Overmind improves performance and efficiency with significantly fewer hardware resources. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_15623 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Overmind NSA: A Unified Neuro-Symbolic Computing Architecture with Approximate Nonlinear Activations and Preemptive Memory Bypass Wang, Weilun Wang, Zirui Li, Wantong Hardware Architecture Neuro-symbolic AI is gaining traction in domains such as large language models, scientific discovery, and autonomous systems due to its ability to combine perception with structured reasoning. However, its deployment is often constrained by high memory demands, diverse computation patterns, and complex hardware requirements. Existing hardware platforms struggle with large on-chip memory overheads, frequent pipeline stalls, limited I/O bandwidth, and inefficient handling of nonlinear operations. To address these key computational bottlenecks, we propose Overmind, a unified neuro-symbolic architecture with cross-layer optimizations. Overmind tackles these core bottlenecks through Padé approximations for universal nonlinear functions, preemptive memory bypass that eliminates costly on-chip caches, and a complete software stack that optimizes model deployment. By reconfiguring the Padé orders for approximating nonlinear functions, we also demonstrate adaptive accuracy-performance scaling. Overmind achieves an energy efficiency of 8.1 TOPS/W and a throughput of 410 GOPS for mixed neuro-symbolic workloads with minimal model accuracy loss. Compared to existing solutions, Overmind improves performance and efficiency with significantly fewer hardware resources. |
| title | Overmind NSA: A Unified Neuro-Symbolic Computing Architecture with Approximate Nonlinear Activations and Preemptive Memory Bypass |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2604.15623 |