# Synaptic Activation and Dual Liquid Dynamics for Interpretable Bio-Inspired Models

arXiv: https://arxiv.org/abs/2602.13017 · HTML: https://mlsanity.com/paper/2602.13017/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 19 references that are in the map.
- First posted on arXiv 13 Feb 2026.

## Builds on (most cited first, 11 of 11)

- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets](https://mlsanity.com/paper/2201.02177/) (2022)
- [Liquid Time-constant Networks](https://mlsanity.com/paper/2006.04439/) (2020)
- [VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles](https://mlsanity.com/paper/2111.12083/) (2021)
- [Minimal Gated Unit for Recurrent Neural Networks](https://mlsanity.com/paper/1603.09420/) (2016)
- [VisualBackProp: efficient visualization of CNNs](https://mlsanity.com/paper/1611.05418/) (2016)
- [Parallelization of Non-linear State-Space Models: Scaling Up Liquid-Resistance Liquid-Capacitance Networks for Efficient Sequence Modeling](https://mlsanity.com/paper/2505.21717/) (2025)
- [Liquid Resistance Liquid Capacitance Networks](https://mlsanity.com/paper/2403.08791/) (2024)
- [Are All Vision Models Created Equal? A Study of the Open-Loop to Closed-Loop Causality Gap](https://mlsanity.com/paper/2210.04303/) (2022)
- [Learning with Chemical versus Electrical Synapses -- Does it Make a Difference?](https://mlsanity.com/paper/2401.08602/) (2023)

## Abstract (arXiv)

In this paper, we present a unified framework for various bio-inspired models to better understand their structural and functional differences. We show that liquid-capacitance-extended models lead to interpretable behavior even in dense, all-to-all recurrent neural network (RNN) policies. We further demonstrate that incorporating chemical synapses improves interpretability and that combining chemical synapses with synaptic activation yields the most accurate and interpretable RNN models. To assess the accuracy and interpretability of these RNN policies, we consider the challenging lane-keeping control task and evaluate performance across multiple metrics, including turn-weighted validation loss, neural activity during driving, absolute correlation between neural activity and road trajectory, saliency maps of the networks' attention, and the robustness of their saliency maps measured by the structural similarity index.