mlsanityData to 7 Oct 2026

Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence

arXiv 2602.12851 ↗First posted 13 Feb 2026cs.NI cs.AI cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

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 primitives can achieve high-fidelity inference within the resource envelope of commodity programmable switches.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 22 of its 50 references that are in the map. First posted on arXiv 13 Feb 2026.

Builds on

22 of 50 references.

References · 22

  1. 2018Efficient Attention: Attention with Linear Complexitiescited by 366
  2. 2019Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoningcited by 149
  3. 2018Kitsune: An Ensemble of Autoencoders for Online Network Intrusion Detectioncited by 121
  4. 2024Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systemscited by 94
  5. 2015A Sensitivity Analysis of (and Practitioners' Guide to) Convolutional Neural Networks for Sentence Classificationcited by 92
  6. 2022Deep Differentiable Logic Gate Networkscited by 67
  7. 2021Skyformer: Remodel Self-Attention with Gaussian Kernel and Nystr"om Methodcited by 61
  8. 2023KDEformer: Accelerating Transformers via Kernel Density Estimationcited by 58
  9. 2024Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformerscited by 51
  10. 2021Stable, Fast and Accurate: Kernelized Attention with Relative Positional Encodingcited by 50
  11. 2024Convolutional Differentiable Logic Gate Networkscited by 37
  12. 2024HardNet: Hard-Constrained Neural Networks with Universal Approximation Guaranteescited by 29
10 more
  1. 2021Energon: Towards Efficient Acceleration of Transformers Using Dynamic Sparse Attentioncited by 26
  2. 2013Programming Protocol-Independent Packet Processorscited by 21
  3. 2025AutoNeural: Co-Designing Vision-Language Models for NPU Inferencecited by 17
  4. 2024Brain-on-Switch: Towards Advanced Intelligent Network Data Plane via NN-Driven Traffic Analysis at Line-Speedcited by 5
  5. 2021An Exhaustive Survey on P4 Programmable Data Plane Switches: Taxonomy, Applications, Challenges, and Future Trendscited by 2
  6. 2025Quark: Implementing Convolutional Neural Networks Entirely on Programmable Data Planecited by 2
  7. 2025Pegasus: A Universal Framework for Scalable Deep Learning Inference on the Dataplanecited by 2
  8. 2025Real-Time In-Network Machine Learning on P4-Programmable FPGA SmartNICs with Fixed-Point Arithmetic and Taylorcited by 2
  9. 2025FENIX: Enabling In-Network DNN Inference with FPGA-Enhanced Programmable Switchescited by 2
  10. 2026SIA: Symbolic Interpretability for Anticipatory Deep Reinforcement Learning in Network Controlcited by 1

Built on by

Not cited yet.