Neural & Evolutionary Computing
Research on neural network architectures, evolutionary algorithms, and bio-inspired computing.
cs.NE · 188 papersSolve the Loop: Attractor Models for Language and Reasoning
Attractor Models introduce a stable, efficient fixed-point refinement method for iterative Transformers, significantly boosting performance in language and reasoning tasks.
A Family of Quaternion-Valued Differential Evolution Algorithms for Numerical Function Optimization
This paper introduces Quaternion-Valued Differential Evolution (QDE) algorithms, showing improved convergence and performance for numerical optimization.
Black-Box Optimization of Mixed Binary-Continuous Variables: Challenges and Opportunities in Evolutionary Model Merging
This paper surveys evolutionary model merging and formally characterizes data flow space merging as a complex black-box optimization problem.
Graph-Grounded Optimization: Rao-Family Metaheuristics, Classical OR, and SLM-Driven Formulation over Knowledge Graphs
Proposes graph-grounded optimization, sourcing problem variables and constraints from knowledge graphs, and evaluates it on diverse real-world problems.
Scaling Laws and Tradeoffs in Recurrent Networks of Expressive Neurons
ELM Networks demonstrate optimal resource allocation in recurrent networks, favoring more complex neurons as scale increases, challenging simple-unit defaults.
Multi-Timescale Conductance Spiking Networks: A Sparse, Gradient-Trainable Framework with Rich Firing Dynamics for Enhanced Temporal Processing
Multi-timescale conductance SNNs offer rich dynamics, sparse activity, and direct gradient training, outperforming SOTA in temporal processing.
Self-organized MT Direction Maps Emerge from Spatiotemporal Contrastive Optimization
A spatiotemporal TDANN model, trained with self-supervised learning, spontaneously generates brain-like direction maps in the visual cortex.
Leveraging Non-Equilibrium ECRAM Dynamics for Short-Term Plasticity in Neuromorphic Circuits
This paper leverages non-equilibrium ECRAM dynamics to efficiently implement short-term plasticity and temporal computation in neuromorphic circuits.
On the Impact of Crossover in Many-Objective Optimization: A Runtime Analysis of NSGA-III
This paper theoretically analyzes NSGA-III, showing crossover significantly speeds up optimization on many-objective problems like m-OJZJ.
Decomposing Evolutionary Mixture-of-LoRA Architectures: The Routing Lever, the Lifecycle Penalty, and a Substrate-Conditional Boundary
This paper decomposes an evolutionary Mixture-of-LoRA system, finding that router improvements, not the evolutionary lifecycle, drive performance gains.
Energy-Efficient Implementation of Spiking Recurrent Cells on FPGA
This paper presents an energy-efficient FPGA accelerator for Spiking Recurrent Cell (SRC) neural networks, balancing biological plausibility and hardware cost.
A Theory of Multilevel Interactive Equilibrium in NeuroAI
MIE is a new game-theoretic framework for NeuroAI, extending Nash equilibrium to intelligent systems with internal computation, partial observability, and bounded rationality.
Causal Explanations from the Geometric Properties of ReLU Neural Networks
This paper generates accurate causal explanations for ReLU neural networks by leveraging their geometric properties, improving interpretability.
Meta-Black-Box Optimization Can Do Search Guidance for Expensive Constrained Multi-Objective Optimization
MetaSG-SAEA introduces a bi-level MetaBBO framework that provides search guidance for expensive constrained multi-objective optimization problems.
Prospective Compression in Human Abstraction Learning
Humans learn abstractions by anticipating future tasks, a "prospective compression" strategy superior to retrospective methods in non-stationary environments.
Frequency Matching in Spiking Neural Networks for mmWave Sensing
This paper introduces frequency matching in SNNs for mmWave sensing, improving accuracy and energy efficiency by aligning LIF dynamics with signal frequencies.
Parameter-Efficient Neuroevolution for Diverse LLM Generation: Quality-Diversity Optimization via Prompt Embedding Evolution
QD-LLM uses neuroevolution to evolve prompt embeddings, enabling diverse and high-quality LLM outputs without fine-tuning.
EvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient Descent
EvoPref, a multi-objective evolutionary algorithm, discovers diverse LLM alignments, overcoming preference collapse in gradient-based methods.
LEVI: Stronger Search Architectures Can Substitute for Larger LLMs in Evolutionary Search
LEVI is an evolutionary search framework that leverages stronger architectures to substitute for larger LLMs, drastically cutting costs while improving performance.
Neuromorphic Reinforcement Learning for Quadruped Locomotion Control on Uneven Terrain
This paper introduces a neuromorphic reinforcement learning framework using equilibrium propagation for quadruped locomotion on uneven terrain, enabling on-robot adaptation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.