archive
Every paper Pith has read. Search by title, abstract, or pith.
554 papers in cs.NE · page 9
-
Concatenated codes boost VSA recovery with formal guarantees
Efficient Vector Symbolic Architectures from Histogram Recovery
-
AI agent rediscovers best math solutions and finds improvements
Mathematical exploration and discovery at scale
-
Neuromorphic chip learns new classes with 113x lower latency
Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network
-
Co-evolving belief and intent boosts driving planner performance
Constructing the Umwelt: Cognitive Planning through Belief-Intent Co-Evolution
-
LLMs from different families share equivalent sinusoidal number structures
Language Models Learn Universal Representations of Numbers and Here's Why You Should Care
-
KANs give Kolmogorov-inspired structure as alternative to MLPs
A Practitioner's Guide to Kolmogorov-Arnold Networks
-
Symbolic emulators match numerical cosmology to 0.001 percent accuracy
Symbolic Emulators for Cosmology: Accelerating Cosmological Analyses Without Sacrificing Precision
-
Barnacle nauplii inspire anchoring method for AUV swarms
Nauplius Optimisation for Autonomous Hydrodynamics
-
Co-design view advances search for spiking neural network architectures
Spiking Neural Network Architecture Search: A Survey
-
Redundant physics terms cut neural parameter errors tenfold
APRIL: Auxiliary Physically-Redundant Information in Loss -- A physics-informed framework for parameter estimation with a gravitational-wave case study
-
Resource-poor foragers aggregate faster through neural hidden states
Emergence of Internal State-Modulated Swarming in Multi-Agent Patch Foraging System
-
Semantic knowledge steers innovation beyond random variation
Semantic knowledge guides innovation and drives cultural evolution
-
Path integrals give exact correlation stats for nonlinear nets
Statistics of correlations in nonlinear recurrent neural networks
-
SpikingMamba cuts LLM energy use by 4.76x with small accuracy gap
SpikingMamba: Towards Energy-Efficient Large Language Models via Knowledge Distillation from Mamba
-
Neural surrogates speed glider wing optimization 785 times
Neural Surrogate-assisted Glider Wing Design with Stability Analysis and Multi-method Optimization
-
ReLU positivity is W[1]-hard parameterized by dimension
Parameterized Hardness of Zonotope Containment and Neural Network Verification
-
CMOS chip with programmable spiking neurons does reservoir computing
CMOS Implementation of Field Programmable Spiking Neural Network for Hardware Reservoir Computing
-
SQR matches black-box quantile accuracy while staying interpretable
Symbolic Quantile Regression for the Interpretable Prediction of Conditional Quantiles
-
Symbolic regression predicts any conditional quantile
Symbolic Quantile Regression for the Interpretable Prediction of Conditional Quantiles
-
Feedback regulation cuts EP convergence time by orders of magnitude
Toward Practical Equilibrium Propagation: Brain-inspired Recurrent Neural Network with Feedback Regulation and Residual Connections
-
Adjoint pullbacks yield coherent neuron explanations in deep nets
Pulling Back the Curtain on Deep Networks
-
Regularization trims popularity bias in graph recommenders
PBiLoss: Popularity-Aware Regularization to Improve Fairness in Graph-Based Recommender Systems
-
Brain physiology supplies design principles for next AGI hardware
Bridging Brains and Machines: A Unified Frontier in Neuroscience, Artificial Intelligence, and Neuromorphic Systems
-
Marker anchors stabilize coevolution in mixed games
Governing Strategic Dynamics: Equilibrium Stabilization via Divergence-Driven Control
-
Dilated groups give 5.8x speedup to masked diffusion decoding
Plan for Speed: Dilated Scheduling for Masked Diffusion Language Models
-
PaceLLM extends LLM context to 200K tokens with brain-like memory
PaceLLM: Brain-Inspired Large Language Models for Long-Context Understanding
-
Residual refinements drive neural nets to machine precision
HiPreNets: High-Precision Neural Networks through Progressive Training
-
Neural loss landscapes hide flat channels to infinity
Flat Channels to Infinity in Neural Loss Landscapes
-
Binary ANN training modeled as Darwiche-Pearl belief revision
Machine Learning as Iterated Belief Change a la Darwiche and Pearl
-
AI agent finds 4x4 matrix multiplication using 48 multiplications
AlphaEvolve: A coding agent for scientific and algorithmic discovery
-
Swarm optimizers cut QAOA approximation gaps versus Adam and SPSA
Benchmarking Swarm Optimization Algorithms for Parameter Initialization in the Quantum Approximate Optimization Algorithm
-
Random features train Hamiltonian graph networks 150-600x faster
Rapid training of Hamiltonian graph networks using random features
-
SiLIF models set new SOTA among spiking neurons on speech tasks
SiLIF: Structured State Space Model Dynamics and Parametrization for Spiking Neural Networks
-
Review finds metaheuristics link to ML in IoT security
A systematic review of metaheuristics-based and machine learning-driven intrusion detection systems in IoT
-
Attractor networks self-orthogonalize via free energy minimization
Self-orthogonalizing attractor neural networks emerging from the free energy principle
-
Frozen backprop cuts weight transport in SNN training
Frozen Backpropagation: Relaxing Weight Symmetry in Deep Spiking Neural Networks
-
Diffusion repair turns evolutionary crossovers into valid 3D molecules
Diffusion-based Evolutionary Optimization for 3D Multi-Objective Molecular Generation
-
Neural network learns robot skills and detects faults without labels
Neural Associative Skill Memories for safer robotics and modelling human sensorimotor repertoires
-
LLMs plan multi-step molecule synthesis routes
LLM-Augmented Chemical Synthesis and Design Decision Programs
-
Personalized AI adapts foundation models without sharing data
A Survey on Foundation Models for Personalized Federated Intelligence
-
GAME alternates sides to reveal arms races in adversarial QD
Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites
-
Spiking networks adapt to shifts by modulating thresholds
Threshold Modulation for Online Test-Time Adaptation of Spiking Neural Networks
-
Z-score filter improves sharpness-aware minimization accuracy
Sharpness-Aware Minimization with Z-Score Gradient Filtering
-
Neural selector picks best agent 86 percent of the time
Neural Orchestration for Multi-Agent Systems: A Deep Learning Framework for Optimal Agent Selection in Multi-Domain Task Environments
-
Stochastic update speeds NSGA-III exponentially on hard problems
Runtime Analyses of NSGA-III on Many-Objective Problems: Provable Exponential Speedup via Stochastic Population Update
-
Kernel scaling yields linear capacity in Hopfield networks
Quantitative Attractor Analysis of High-Capacity Kernel Hopfield Networks
-
Tightest Neighbors form clusters from multi-density streams
TNStream: Applying Tightest Neighbors to Micro-Clusters to Define Multi-Density Clusters in Streaming Data
-
Transformer in RL lifts SFC partitioning acceptance rates
Transformer-Empowered Actor-Critic Reinforcement Learning for Sequence-Aware Service Function Chain Partitioning
-
LLM datetime primitives unreliable but learnable via synthetic data
PRIMETIME : Limits of LLMs in Temporal Primitives
-
ALDA4Rec outperforms baselines in accuracy on four datasets
Adaptive Long-term Embedding with Denoising and Augmentation for Recommendation