![[2602.12851] Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.12851] Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence
Summary
The paper presents Chimera, a framework that integrates neuro-symbolic attention mechanisms into programmable dataplanes, enhancing traffic analysis with trustworthy inference while adhering to hardware constraints.
Why It Matters
Chimera addresses the critical need for reliable and efficient traffic analysis in networking by combining neural computations with symbolic constraints. This approach not only improves performance but also ensures predictable and auditable behavior, which is essential for modern network applications.
Key Takeaways
- Chimera enables trustworthy inference in programmable dataplanes.
- The framework combines kernelized attention with symbolic guarantees.
- Empirical results demonstrate high-fidelity inference under resource constraints.
- A hardware-aware mapping protocol ensures stable operation.
- The approach is significant for enhancing network traffic analysis.
Computer Science > Networking and Internet Architecture arXiv:2602.12851 (cs) [Submitted on 13 Feb 2026] Title:Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence Authors:Rong Fu, Wenxin Zhang, Xiaowen Ma, Kun Liu, Wangyu Wu, Ziyu Kong, Jia Yee Tan, Tailong Luo, Xianda Li, Zeli Su, Youjin Wang, Yongtai Liu, Simon Fong View a PDF of the paper titled Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence, by Rong Fu and 12 other authors View PDF HTML (experimental) Abstract:Deploying expressive learning models directly on programmable dataplanes promises line-rate, low-latency traffic analysis but remains hindered by strict hardware constraints and the need for predictable, auditable behavior. Chimera introduces a principled framework that maps attention-oriented neural computations and symbolic constraints onto dataplane primitives, enabling trustworthy inference within the match-action pipeline. Chimera combines a kernelized, linearized attention approximation with a two-layer key-selection hierarchy and a cascade fusion mechanism that enforces hard symbolic guarantees while preserving neural expressivity. The design includes a hardware-aware mapping protocol and a two-timescale update scheme that together permit stable, line-rate operation under realistic dataplane budgets. The paper presents the Chimera architecture, a hardware mapping strategy, and empirical evidence showing that neuro-symbolic attention primi...
