A taxonomy of SNN training algorithms is presented with the release of NeuroTrain, an open benchmarking framework for reproducible comparisons across datasets and architectures.
Canonical reference
PLOS ONE7(6)
Canonical reference. 86% of citing Pith papers cite this work as background.
38
Pith papers citing it
Background
86% of classified citations
citation-role summary
background 6
other 1
citation-polarity summary
representative citing papers
Discourse among AI agents on Moltbook is largely determined by architectural constraints like context windows and identity files, appearing as social learning but actually short-horizon contextual conditioning.
TikTok formally complies with DSA rules against profiling minors but delivers 5-8 times stronger interest-based targeting through undisclosed influencer and promotional content.
Optimal resilience configurations exist for networks with three or more nodes, with primary links scaling as O(N^{-1}) and background links as O(N^{-2}) with logarithmic corrections.
DLCM is a lattice-based stochastic solver for multicellular systems that introduces elliptic projection to couple cell pressure with population curvature for surface tension effects.
CDER-SME provides 1,963 expert-annotated Event-RGB micro-expression samples from 92 subjects under multi-level stress, with a hardware-agnostic alignment pipeline and a multimodal baseline showing fusion gains.
OlfactProfile shows structured field-wise profile conditioning improves odor prediction from audiovisual content over naive methods, with gains on background and emotion odors in a new 1,350-clip benchmark using the OAR fusion module.
Frontier LLMs exhibit frame-dependent Bitcoin preferences; a sparse autoencoder feature in Gemma 3 causally modulates portfolio shares by 4-5 percentage points when amplified or suppressed.
Existing visual attribution methods often fail to identify the visual evidence used by LVLMs in chest X-ray reasoning, while MedFocus using unbalanced optimal transport and targeted interventions substantially outperforms them across multiple models and settings.
Unifies multiple nestedness measures in higher-order networks via the encapsulation DAG and demonstrates that nested structure is prevalent in social systems while its absence indicates mesoscale organization.
AI coding tools divide into collaborators that initiate most PRs and assistants that support human-led ones, yet humans retain merge authority across all five tools examined.
A new Bayesian multiscale framework with cut inference jointly models heterogeneous viral load trajectories and household transmission, recovering parameters without bias on simulated data when viral sampling is frequent.
The adaptive bounded-rationality model anticipates hazardous takeovers with better coverage and lead time than baselines while aligning inferred parameters with eye-tracking metrics.
Glide-in-Place uses per-foot pressure sensing for continuous differential-drive VR locomotion, outperforming seated walking-in-place on speed and fatigue while matching joystick performance on simulator sickness in a 16-person study.
A Pretty Good Measurement classifier reformulates multi-class radiomics as quantum state discrimination and achieves competitive performance on NSCLC subtyping and PCa risk tasks.
Survey of 868 scientific programmers shows generative AI adoption is highest among the inexperienced, who prefer conversational tools, and perceived productivity correlates most with volume of accepted generated code rather than validation practices.
A new algorithm simulates exact trajectories of bursty PDMP models for arbitrary interacting genes at lower cost than SSA, with a toggle-switch example showing bimodal expression arises from interaction-driven burst frequencies rather than bursting alone.
Coverage bias in GPS mobility data is source-specific and spatially dependent, with Facebook showing more even coverage than multi-app data and different dominant drivers for each.
IUU+DB is an LLM-driven pipeline that classifies documents, extracts structured incident data on IUU+ activities, and supports deduplication and trend analysis.
SSL speech representations outperform hand-crafted features at lower cognitive hierarchy levels but reverse for MCI classification, with greater response freedom in tasks linked to performance dilution at higher levels and structured tasks showing the opposite pattern.
Human-human interaction in RPS can produce higher Lempel-Ziv complexity sequences than RNG opponents via sensitivity to recent frequency biases, most evident in low-entropy opponent states.
Non-unique time arising from event-driven order flow points to a foundational market incompleteness beyond usual no-arbitrage assumptions.
A computational MSV model built from short-video multimodal features positively predicts sensory engagement but shows an inverted-U relationship with behavioral engagement, validated on large unseen platform datasets.
Preprints correlate with 19% higher citations, software sharing with 13.5%, data sharing with 14.3%, and open access with 8.6% in a citation prediction model applied to the French Open Science Monitor dataset.
citing papers explorer
-
NeuroTrain: Surveying Local Learning Rules for Spiking Neural Networks with an Open Benchmarking Framework
A taxonomy of SNN training algorithms is presented with the release of NeuroTrain, an open benchmarking framework for reproducible comparisons across datasets and architectures.
-
What Do AI Agents Talk About? Discourse and Architectural Constraints in the First AI-Only Social Network
Discourse among AI agents on Moltbook is largely determined by architectural constraints like context windows and identity files, appearing as social learning but actually short-horizon contextual conditioning.
-
The DSA's Blind Spot: Algorithmic Audit of Advertising and Minor Profiling on TikTok
TikTok formally complies with DSA rules against profiling minors but delivers 5-8 times stronger interest-based targeting through undisclosed influencer and promotional content.
-
On the existence of Ulanowicz's optimal structural resilience in complex networks
Optimal resilience configurations exist for networks with three or more nodes, with primary links scaling as O(N^{-1}) and background links as O(N^{-2}) with logarithmic corrections.
-
DLCM: a versatile multi-level solver for heterogeneous multicellular systems
DLCM is a lattice-based stochastic solver for multicellular systems that introduces elliptic projection to couple cell pressure with population curvature for surface tension effects.
-
CDER-SME: A Cross-Device Event-RGB Micro-Expression Dataset under Multi-Level Stress Induction
CDER-SME provides 1,963 expert-annotated Event-RGB micro-expression samples from 92 subjects under multi-level stress, with a hardware-agnostic alignment pipeline and a multimodal baseline showing fusion gains.
-
OlfactProfile: Profile-Conditioned Odor Prediction from Audiovisual Content
OlfactProfile shows structured field-wise profile conditioning improves odor prediction from audiovisual content over naive methods, with gains on background and emotion odors in a new 1,350-clip benchmark using the OAR fusion module.
-
Auditing Asset-Specific Preferences in Financial Large Language Models: Evidence from Bitcoin Representations and Portfolio Allocation
Frontier LLMs exhibit frame-dependent Bitcoin preferences; a sparse autoencoder feature in Gemma 3 causally modulates portfolio shares by 4-5 percentage points when amplified or suppressed.
-
Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models
Existing visual attribution methods often fail to identify the visual evidence used by LVLMs in chest X-ray reasoning, while MedFocus using unbalanced optimal transport and targeted interventions substantially outperforms them across multiple models and settings.
-
The nestedness of higher-order networks
Unifies multiple nestedness measures in higher-order networks via the encapsulation DAG and demonstrates that nested structure is prevalent in social systems while its absence indicates mesoscale organization.
-
Collaborator or Assistant? How AI Coding Agents Partition Work Across Pull Request Lifecycles
AI coding tools divide into collaborators that initiate most PRs and assistants that support human-led ones, yet humans retain merge authority across all five tools examined.
-
Bayesian inference for disease transmission models informed by viral dynamics
A new Bayesian multiscale framework with cut inference jointly models heterogeneous viral load trajectories and household transmission, recovering parameters without bias on simulated data when viral sampling is frequent.
-
Adaptive Bounded-Rationality Modeling of Early-Stage Takeover in Shared-Control Driving
The adaptive bounded-rationality model anticipates hazardous takeovers with better coverage and lead time than baselines while aligning inferred parameters with eye-tracking metrics.
-
Glide-in-Place: Foot-Steered Differential-Drive for Hands-Free VR Locomotion
Glide-in-Place uses per-foot pressure sensing for continuous differential-drive VR locomotion, outperforming seated walking-in-place on speed and fatigue while matching joystick performance on simulator sickness in a 16-person study.
-
Pretty Good Measurement for Radiomics: A Quantum-Inspired Multi-Class Classifier for Lung Cancer Subtyping and Prostate Cancer Risk Stratification
A Pretty Good Measurement classifier reformulates multi-class radiomics as quantum state discrimination and achieves competitive performance on NSCLC subtyping and PCa risk tasks.
-
A survey of generative AI adoption and perceived productivity among scientists who program
Survey of 868 scientific programmers shows generative AI adoption is highest among the inexperienced, who prefer conversational tools, and perceived productivity correlates most with volume of accepted generated code rather than validation practices.
-
Efficient stochastic simulation of gene regulatory networks using hybrid models of transcriptional bursting
A new algorithm simulates exact trajectories of bursty PDMP models for arbitrary interacting genes at lower cost than SSA, with a toggle-switch example showing bimodal expression arises from interaction-driven burst frequencies rather than bursting alone.
-
One country, multiple portraits: representativeness in GPS-based mobility data is source-specific and spatially dependent
Coverage bias in GPS mobility data is source-specific and spatially dependent, with Facebook showing more even coverage than multi-app data and different dominant drivers for each.
-
IUU+DB: Tracking Illegal, Unreported, and Unregulated Fishing, Seafood Fraud, and Labor Abuse through LLM-driven Information Extraction
IUU+DB is an LLM-driven pipeline that classifies documents, extracts structured incident data on IUU+ activities, and supports deduplication and trend analysis.
-
Beyond Binary: Speech Representations Across the Cognitive Score Hierarchy
SSL speech representations outperform hand-crafted features at lower cognitive hierarchy levels but reverse for MCI classification, with greater response freedom in tasks linked to performance dilution at higher levels and structured tasks showing the opposite pattern.
-
Toward an Origin of Human Randomness: Interaction-Driven Enhancement in the Rock-Paper-Scissors Game
Human-human interaction in RPS can produce higher Lempel-Ziv complexity sequences than RNG opponents via sensitivity to recent frequency biases, most evident in low-entropy opponent states.
-
Non-unique time and market incompleteness
Non-unique time arising from event-driven order flow points to a foundational market incompleteness beyond usual no-arbitrage assumptions.
-
A Computational Model of Message Sensation Value in Short Video Multimodal Features that Predicts Sensory and Behavioral Engagement
A computational MSV model built from short-video multimodal features positively predicts sensory engagement but shows an inverted-U relationship with behavioral engagement, validated on large unseen platform datasets.
-
An analysis of the effects of open science indicators on citations in the French Open Science Monitor
Preprints correlate with 19% higher citations, software sharing with 13.5%, data sharing with 14.3%, and open access with 8.6% in a citation prediction model applied to the French Open Science Monitor dataset.
-
Deadeye Visualization Revisited: Investigation of Preattentiveness and Applicability in Virtual Environments
Deadeye preserves preattentiveness in VR with heterogeneous distractors (accuracies above 90%) and shows applicability to volume rendering workflows.
-
Learning to model pediatric asthma exacerbation from multiple risk factors: a case study in coastal Virginia
A case study develops a sparse dictionary learning approach to model pediatric asthma exacerbations from multiple risk factors and reports consensus on relative risks across statistical and machine learning models.
-
Building Digital Societies as Ecosystems: How Recognition and Repeat Relationships Sustain Cross-Community Work in Open Source
Cross-boundary collaboration in open source is sustained by a thin carrier layer of contributors and repeat relationships that increase pull request acceptance rates from 42% to 87%.
-
Active Inference: A method for Phenotyping Agency in AI systems?
Active inference offers a variational way to phenotype agency in AI systems by measuring empowerment in generative models via a T-maze paradigm.
-
Climate-Driven Dengue Forecasting in Bangladesh: Division-Specific Feature-Set Design and Lag Structure
ANN with one climate feature set gives the lowest error for Dhaka dengue forecasts while SARIMAX with a second set performs best for Barishal, with rainfall and humidity showing positive associations at 1-2 month lags.
-
A Framework for Longitudinal Health AI Agents
Proposes a multi-layer framework and agent architecture that operationalizes adaptation, coherence, continuity, and agency for longitudinal health AI agents.
-
A Gradient Boosted Mixed-Model Machine Learning Framework for Vessel Speed in the U.S. Arctic
A two-stage gradient boosted model with random effects predicts Arctic vessel movement probability (AUC 0.85) and conditional positive speed (77% out-of-fold variance explained), highlighting distance to coast and bathymetric depth as dominant factors.
-
Analysis and Control of Acoustic Emissions from Marine Energy Converters
Simulation study finds direct-drive generators eliminate mechanical tonal noise in tidal converters while detuning MPPT by a factor of 1.2 cuts mammal risk with 3.58% energy yield loss.
-
Autoencoding sensory substitution
Deep recurrent autoencoders convert images to shortened audio signals that incorporate hearing models, enabling above-chance hand posture discrimination and object reaching after a few hours of training instead of months.
-
GPU-Accelerated Deep Learning for Heatwave Prediction and Urban Heat Risk Assessment
ConvLSTM with mixed loss achieves MAE 0.2293, RMSE 0.3089, R2 0.8877 for next-day temperature prediction in Sarajevo and supports generation of urban heat-risk maps.
-
Exploring the non-convexity in machine learning using quantum-inspired optimization
A quantum-inspired global search method called QIEO outperforms traditional solvers in recovering sparse structures and robust fitting by maintaining a broad view of possible solutions.
-
Challenger at MultiPRIDE: Is It Hate Speech or Reclaimed?
Applies embeddings, Cleanlab noise filtering, and MLP classification to achieve robust performance on imbalanced MultiPride data for distinguishing hate speech from reclaimed language.
- Alternate states and intermingledness in complex high-dimensional systems
- "I Said Things I Needed to Hear Myself": Peer Support as an Emotional, Organisational, and Sociotechnical Practice in Singapore