The ANTShapes simulation tool output has been validated as suitable for creating and labeling event-based vision datasets
Research into event-based object classification methods is hindered by the lack of high-quality vision datasets
Event-based object classification using Spiking Neural Networks on neuromorphic devices can solve issues of size, weight, power consumption, security, latency, and connectivity present in conventional frame-based computer vision approaches
Conventional frame-based computer vision approaches have practical flaws including size, weight, power consumption constraints, security concerns with cloud computation, transmission latency, and connectivity requirements
Relying solely on lexical similarity between source code and bug reports is often insufficient for bug localization due to the natural language nature of bug descriptions
The bug localization recommender system successfully identified buggy classes for 88.5% of bug reports within the top 10 recommendations and 94% within the top 20
The multi-objective bug localization model demonstrates adaptability across programming languages, validated on both Java and Kotlin projects
A class-level automated multi-objective search-based system using SPEA-2 achieved higher precision and recall than NSGA-II and MOEA/D for bug localization
Physical reservoir computing systems can implement causal information filtering through asymmetric coupling anisotropy, creating deterministic upstream-to-downstream information flow
The intrinsic causality of reservoir topology provides a robust framework for autonomous reliability and fault-tolerant physical intelligence
Local phase transitions in physical reservoir computing networks can purge anomalous information while preserving computational integrity of remaining nodes
Asymmetric coupling anisotropy in physical reservoir computing allows for selective amplification of semantic drifts, triggering macroscopic saddle-node bifurcation as a physical interlock before global computational failure
Route-aware, activity-weighted multicast hypergraph mapping provides a better abstraction for mapping spiking neural networks onto neuromorphic many-core platforms than traditional graph partitioning with pairwise placement costs
M-HySMap reduces routed multicast hops by 10.6-19.6% over Activity+QAP and 19.7-41.1% over Edge+QAP across a 115-job evidence suite on Potjans-inspired recurrent SNNs and mesh NoCs from 4x4 to 6x6, plus a 7x7 stress case
Confidence-sensitive transmission amplifies social error in reinforcement learning agents, while confidence-dependent private learning stabilizes it
Incremental updates in M-HySMap accelerate refinement by 4.7-12.7x while matching full recomputation to numerical precision
The paper resolves an open problem posted at COLT'25 regarding parameterized complexity of zonotope norm maximization
20-70% of sequenced genes in non-model organisms lack characterized functions
The authors used LLMs as part of their research process
Computing Lipschitz constants is difficult even for shallow ReLU networks