Selected work

Representative Papers

The RAIL Principles for Neurosymbolic AI: Reasoning, Assurances, Interfacing and Learning

Aug 04, 2026

This work addresses the insufficient reliability and trustworthiness of AI systems in high-stakes or data-scarce scenarios by proposing RAIL—a unified design framework for neuro-symbolic AI grounded in four principles: Reasoning, Assurance, Interface, and Learning. Integrating cutting-edge techniques such as physics-informed learning, causal inference, and tool-augmented large language models, RAIL offers engineers actionable design guidelines through neuro-symbolic integration, formal reasoning, and neural-guided search. The framework not only fosters deep synergy between neural and symbolic approaches but also substantially enhances AI system performance in reliability, efficiency, and trustworthiness, demonstrating broad applicability across real-world domains.

0 citationsRead paper

Context-Informed Ship Trajectory Prediction via Conditional Attention

Jul 29, 2026

This work addresses the limitations of existing vessel trajectory prediction methods, which often neglect the directional physical influence of environmental factors and suffer significant performance degradation under sensor outages. To overcome these issues, the authors propose a Transformer-based conditional generative framework that explicitly models the unidirectional modulation of vessel dynamics by environmental conditions through a conditional attention mechanism. Furthermore, a modality masking training strategy is introduced to enhance robustness against unreliable or missing sensor inputs. By integrating Automatic Identification System (AIS) data with ERA5 meteorological reanalysis, the proposed method achieves a 15.4% improvement in prediction accuracy when environmental context is available and reduces fallback error by nearly an order of magnitude during sensor failures.

0 citationsRead paper

Multi-Agent Planning with Spatio-Temporal and Topological Constraints using STL-GO

Jul 29, 2026

This work addresses the challenge of jointly planning spatiotemporal constraints—specifying when and where tasks are executed—and topological constraints—governing agent interaction structures—in multi-agent systems. To this end, the authors propose a unified modeling framework grounded in STL-GO logic, which for the first time incorporates dynamic multi-graph interactions. They develop two complete solution approaches based on Mixed-Integer Programming (MIP) and Satisfiability Modulo Theories (SMT), enabling seamless switching and comparative analysis between the two paradigms. The effectiveness of the proposed method is validated on a multi-UAV search-and-rescue benchmark, demonstrating strong expressiveness and favorable scalability across varying team sizes and levels of time-varying graph complexity.

0 citationsRead paper

AI-interpreted Optical Scattering for Robust and Focal Depth-Aware Imaging

Jul 24, 2026

Conventional imaging typically treats light scattering as a nuisance, overlooking its potential utility. This work demonstrates for the first time that optical scattering can enhance image robustness against pixel loss and encode depth-of-focus information. To investigate this, we introduce the Scattering MNIST dataset, which incorporates varying scattering conditions, and combine physically grounded optical scattering models with a variational autoencoder (VAE) to analyze speckle patterns through an interpretable latent space. Experimental results show that our approach achieves reconstruction accuracy comparable to state-of-the-art deep learning models while simultaneously enabling effective depth-of-focus discrimination and improved robustness to missing pixels.

0 citationsRead paper

A Hardware-Aware Open-Source Framework for Design Space Exploration of Mixed-Signal Spiking Neural Networks

Jul 07, 2026

This work addresses the lack of simulation tools capable of jointly modeling hardware non-idealities in mixed-signal spiking neural networks (SNNs) while enabling system-level design exploration. The authors propose an open-source, hardware-aware SNN simulation framework embedded within PyTorch that, for the first time, integrates LIF, Hodgkin-Huxley (HH), and Axon-Hillock neuron models with non-volatile analog synapses based on floating-gate transistors and ReRAM devices within a unified platform. Crucially, device nonlinearities and non-ideal characteristics are directly incorporated into the training pipeline. The framework supports end-to-end optimization and cross-layer design space exploration, enabling joint evaluation of classification accuracy, silicon area, power consumption, and quantization sensitivity on benchmarks such as N-MNIST, DVS Gesture, and SHD, thereby facilitating multi-objective-constrained search for optimal neuron-synapse configurations.

0 citationsRead paper

Recent publications

Latest Papers

The RAIL Principles for Neurosymbolic AI: Reasoning, Assurances, Interfacing and Learning

Aug 04, 2026

This work addresses the insufficient reliability and trustworthiness of AI systems in high-stakes or data-scarce scenarios by proposing RAIL—a unified design framework for neuro-symbolic AI grounded in four principles: Reasoning, Assurance, Interface, and Learning. Integrating cutting-edge techniques such as physics-informed learning, causal inference, and tool-augmented large language models, RAIL offers engineers actionable design guidelines through neuro-symbolic integration, formal reasoning, and neural-guided search. The framework not only fosters deep synergy between neural and symbolic approaches but also substantially enhances AI system performance in reliability, efficiency, and trustworthiness, demonstrating broad applicability across real-world domains.

0 citationsRead paper

Context-Informed Ship Trajectory Prediction via Conditional Attention

Jul 29, 2026

This work addresses the limitations of existing vessel trajectory prediction methods, which often neglect the directional physical influence of environmental factors and suffer significant performance degradation under sensor outages. To overcome these issues, the authors propose a Transformer-based conditional generative framework that explicitly models the unidirectional modulation of vessel dynamics by environmental conditions through a conditional attention mechanism. Furthermore, a modality masking training strategy is introduced to enhance robustness against unreliable or missing sensor inputs. By integrating Automatic Identification System (AIS) data with ERA5 meteorological reanalysis, the proposed method achieves a 15.4% improvement in prediction accuracy when environmental context is available and reduces fallback error by nearly an order of magnitude during sensor failures.

0 citationsRead paper

Multi-Agent Planning with Spatio-Temporal and Topological Constraints using STL-GO

Jul 29, 2026

This work addresses the challenge of jointly planning spatiotemporal constraints—specifying when and where tasks are executed—and topological constraints—governing agent interaction structures—in multi-agent systems. To this end, the authors propose a unified modeling framework grounded in STL-GO logic, which for the first time incorporates dynamic multi-graph interactions. They develop two complete solution approaches based on Mixed-Integer Programming (MIP) and Satisfiability Modulo Theories (SMT), enabling seamless switching and comparative analysis between the two paradigms. The effectiveness of the proposed method is validated on a multi-UAV search-and-rescue benchmark, demonstrating strong expressiveness and favorable scalability across varying team sizes and levels of time-varying graph complexity.

0 citationsRead paper

AI-interpreted Optical Scattering for Robust and Focal Depth-Aware Imaging

Jul 24, 2026

Conventional imaging typically treats light scattering as a nuisance, overlooking its potential utility. This work demonstrates for the first time that optical scattering can enhance image robustness against pixel loss and encode depth-of-focus information. To investigate this, we introduce the Scattering MNIST dataset, which incorporates varying scattering conditions, and combine physically grounded optical scattering models with a variational autoencoder (VAE) to analyze speckle patterns through an interpretable latent space. Experimental results show that our approach achieves reconstruction accuracy comparable to state-of-the-art deep learning models while simultaneously enabling effective depth-of-focus discrimination and improved robustness to missing pixels.

0 citationsRead paper

A Hardware-Aware Open-Source Framework for Design Space Exploration of Mixed-Signal Spiking Neural Networks

Jul 07, 2026

This work addresses the lack of simulation tools capable of jointly modeling hardware non-idealities in mixed-signal spiking neural networks (SNNs) while enabling system-level design exploration. The authors propose an open-source, hardware-aware SNN simulation framework embedded within PyTorch that, for the first time, integrates LIF, Hodgkin-Huxley (HH), and Axon-Hillock neuron models with non-volatile analog synapses based on floating-gate transistors and ReRAM devices within a unified platform. Crucially, device nonlinearities and non-ideal characteristics are directly incorporated into the training pipeline. The framework supports end-to-end optimization and cross-layer design space exploration, enabling joint evaluation of classification accuracy, silicon area, power consumption, and quantization sensitivity on benchmarks such as N-MNIST, DVS Gesture, and SHD, thereby facilitating multi-objective-constrained search for optimal neuron-synapse configurations.

0 citationsRead paper