All authors
Claude Skills by hiyenwong
github.com/hiyenwong9,934 skills5 installs19,223 views
- Qubo Federated Learning SecurityQUBO-based Byzantine-resilient federated learning methodology. Reformulates client selection in federated learning as Quadratic Unconstrained Binary Optimization solved by quantum annealers. Jointly optimizes over all client subsets to find mutually closest group, outperforming greedy MultiKrum on advanced Byzantine attacks. Use when: QUBO client selection, Byzantine-resilient federated learning, quantum annealing FL, QUBO optimization, federated learning security, MultiKrum improvement, quan...Votes: 0GitHub stars: 3
- Rad 2 Scaling Reinforcement Learning Generator Discriminator FraResearch methodology from paper 'RAD-2: Scaling Reinforcement Learning in a Generator-Discriminator Framework'. arXiv:2604.15308v1. Covers key techniques and approaches for neuroscience research. Activation: rad, 2, scaling, cs.CVVotes: 0GitHub stars: 3
- Random Graph Datacenter NetworksRandom graph-based datacenter network design (RNG) for cost-optimized, fault-tolerant distributed systems. Covers distributed routing protocols exploiting random graph properties, passive optical cabling shuffles, and production deployment patterns. Use when: designing datacenter topologies, optimizing network cost vs performance, implementing distributed routing on non-hierarchical graphs, deploying fault-tolerant network fabrics, or evaluating random graph vs fat-tree tradeoffs.Votes: 0GitHub stars: 3
- Recode Agent Workflow[TODO: Complete and informative explanation of what the skill does and when to use it. Include WHEN to use this skill - specific scenarios, file types, or tasks that trigger it.]Votes: 0GitHub stars: 3
- Recursive Multi Agent SystemsRecursiveMAS framework for scaling multi-agent collaboration through recursive latent-space computation. Enables heterogeneous agents to form collaboration loops via RecursiveLink module with inner-outer loop learning. Activation triggers: recursive multi-agent, RecursiveMAS, latent-space agent collaboration, recursive scaling, agent loop, cross-agent latent transfer.Votes: 0GitHub stars: 3
- Relaxation Informed Surrogate TrainingRelaxation-Informed Training (RIT) for neural network surrogate models enabling exact MILP embedding with reduced binary variables. Activation triggers: surrogate optimization, MILP embedding, neural network relaxation, ReLU pruning, optimization surrogateVotes: 0GitHub stars: 3
- Resolvent Rnn Multi Hop SparsityResolvent-RNN (R-RNN) methodology for constraining multi-hop temporal pathways in recurrent neural networks to achieve temporal sparsity alignment. Use when: (1) analyzing or designing RNN architectures with multi-hop temporal dependencies, (2) studying temporal sparsity in sequence modeling, (3) understanding resolvent-based constraints for recurrent dynamics, (4) improving RNN long-range dependency handling, (5) researching spectral methods for RNN stability and expressivity. Triggers: R-RN...Votes: 0GitHub stars: 3
- Rft Visual Continual LearningUsing Reinforcement Fine-Tuning (RFT/GRPO) to overcome catastrophic forgetting in visual continual learning. Activation triggers: reinforcement fine-tuning continual learning, GRPO visual CL, RL visual continual learning, catastrophic forgetting visual, class-incremental visual learningVotes: 0GitHub stars: 3
- Rhythm Switching Adaptive Time Constants RnnMethodology for analyzing how recurrent neural networks with neuron-specific adaptive time constants switch between multiple frequency band rhythms. Covers rhythm-switching mechanisms, time constant-frequency relationships, and degeneracy of learned solutions. Activation: rhythm switching RNN, adaptive time constants, frequency band switching, RNN neural dynamics, multi-band rhythms, cortical rhythm mechanisms.Votes: 0GitHub stars: 3
- Rim Reasoning Memory Llm Working MemoryRiM (Reasoning in Memory) methodology for unlocking working memory capacity in LLMs via fixed memory blocks, enabling compute-efficient latent reasoning without autoregressive thought generation.Votes: 0GitHub stars: 3
- Rnn Task Degradation AnalysisRNN权重初始化、解的多样性与性能退化分析框架。研究不同初始化如何收敛到不同动力学解，分析网络规模、时间间隔、连接损伤对性能的优雅退化影响。适用于计算神经科学、RNN模型分析、脑皮层建模。触发词：RNN初始化、解多样性、性能退化、网络鲁棒性、优雅退化、weight initialization、degradation analysis、RNN dynamics、graceful degradation。Votes: 0GitHub stars: 3
- Routing Distraction Multimodal MoeRouting analysis and intervention for Multimodal Mixture-of-Experts models. Use when: (1) Debugging vision-language reasoning failures, (2) Analyzing expert routing in MoE architectures, (3) Improving multimodal MoE performance, (4) Understanding cross-modal expert activation. Triggers: mixture-of-experts, MoE routing, multimodal reasoning, vision-language models, expert activation, routing intervention, cross-modal distraction.Votes: 0GitHub stars: 3
- Sample Optimal Gaussian State LearningSample complexity bounds and algorithms for learning bosonic Gaussian quantum states. Use when: (1) analyzing sample requirements for quantum state tomography, (2) designing efficient measurement strategies for Gaussian states, (3) computing sample complexity lower/upper bounds for continuous-variable systems, (4) determining when non-Gaussian measurements are required, (5) optimizing adaptive measurement schemes for quantum state learning. Activation: Gaussian state tomography, sample comple...Votes: 0GitHub stars: 3
- Search E1 Self DistillationSearch-E1 methodology from arXiv:2605.22511 (May 2026). Self-evolution for search-augmented reasoning agents via vanilla GRPO + Offline Self-Distillation (OFSD) with token-level forward KL objective. No external supervision needed. Use when: training search-augmented LLM agents, self-evolution pipelines, GRPO-based reasoning, offline distillation from privileged context.Votes: 0GitHub stars: 3
- Selective Forgetting Agent Memory BiologicalBiologically-inspired selective forgetting framework for LLM agent memory management. Combines hippocampal indexing theory and Ebbinghaus forgetting curve for efficient, secure, and quality-preserving memory pruning. Use for: agent memory optimization, privacy-preserving AI, memory-constrained deployment. Triggers: selective forgetting, memory pruning, agent memory, forgetting mechanism, hippocampal theory.Votes: 0GitHub stars: 3
- Self Organising TransformerSelf-organising transformer architectures that determine their own structure during training. DDCL-INCRT pattern with hierarchical prototype structure. Use when designing adaptive neural architectures, self-organising networks, or architectures that evolve structure. Activation: self-organising transformer, DDCL, adaptive architecture, self-organising network, 自组织架构, prototype learning.Votes: 0GitHub stars: 3
- Self Orthogonalizing Attractor NetworksFormalizes how attractor networks emerge from the free energy principle applied to universal partitioning of random dynamical systems. Results in self-orthogonalizing attractor representations, biologically plausible multi-level Bayesian active inference. Use when: studying attractor dynamics in neural networks, free energy principle applications, Bayesian active inference models, biologically plausible learning, Boltzmann Machine variants, self-organizing neural dynamics. Triggered by: free ...Votes: 0GitHub stars: 3
- Self Policy Distillation SpdSelf-Policy Distillation (SPD) methodology from arXiv:2605.22675 (May 2026). Capability-selective self-distillation by extracting low-rank subspace from gradients on correctness-defining tokens, projecting KV activations during self-generation, and fine-tuning with standard NTP loss. No external signals needed. Use when: self-distillation, self-improvement of LLMs, capability-selective training, activation subspace projection, generalizable self-training.Votes: 0GitHub stars: 3
- Self Supervised Local Learning HierarchyBiologically plausible local self-supervised learning rules that learn hidden hierarchical data structure as efficiently as supervised backprop. Demonstrates that Direct Feedback Alignment (DFA) methods fail on hierarchical tasks due to input-specific masking. Use for biologically plausible learning algorithms, local plasticity rules, self-supervised representation learning.Votes: 0GitHub stars: 3
- Self Supervised Local Learning RhmLayerwise self-supervised local learning rules for deep networks on the Random Hierarchy Model (RHM). Direct feedback approximations fail due to input-specific masking nonlinearity; layerwise contrastive/non-contrastive self-supervised rules succeed and match backprop data efficiency while being cortex-compatible. Use when: designing biologically plausible learning rules, local learning algorithms, self-supervised contrastive learning, solving weight transport problem, hierarchical structure ...Votes: 0GitHub stars: 3
- Sensation Modulating NetworkSensation Modulating Network (SMN) architecture for embodied cognition - haltability-based framework reconciling cognitivism and 4E approaches through opponent dynamics. Use when studying embodied cognitive architecture, intentional directedness without modules, self/world distinction as structural feature, or generativity from autonomic regularity. Keywords: haltability, opponent dynamics, sensation modulators, SMAP, cognitivism vs 4E, embodied architecture, intentional directedness.Votes: 0GitHub stars: 3
- Sequential Chaotic Oscillations Ei NetworksSequential chaotic oscillations (SCOs) in excitatory-inhibitory threshold-linear networks - dynamical mechanism for sequential metastability in brain dynamics. Activation: sequential metastability, chaotic itinerancy, E-I oscillation, SCO, threshold-linear network, brain dynamics, metastable states.Votes: 0GitHub stars: 3
- Sleep Like Consolidation LlmSleep-like consolidation mechanism for LLMs that converts recent context into persistent fast weights before clearing KV cache, enabling long-horizon reasoning with preserved inference latency.Votes: 0GitHub stars: 3
- Sparse Gradient Plasticity梯度突触可塑性稀疏实现方法论。实现真正稀疏且通用的梯度突触可塑性规则，保持在线学习能力。适用于脉冲神经网络、在线学习、突触可塑性研究。触发词：突触可塑性、梯度下降、稀疏实现、在线学习、sparse plasticity、gradient-based、online learning。Votes: 0GitHub stars: 3
- Spec Driven Agent ArchitectureWorkflow and architecture patterns for building robust AI agents using Specs, Contracts, and Repository patterns.Votes: 0GitHub stars: 3
- Specificity Aware Federated Graph Learning特异性感知联邦图学习框架(SFGL)。在保护数据隐私的前提下进行多站点fMRI协作训练，通过共享分支和个性化分支平衡知识共享与站点特异性保持。适用于多中心脑疾病识别、联邦学习、rs-fMRI分析。触发词：联邦学习、图神经网络、fMRI分析、多站点、数据隐私、federated learning、GNN、multi-site、brain disorder identification。Votes: 0GitHub stars: 3
- Speculative Decoding OptimizationOptimize LLM inference using speculative decoding with KV cache compression. Covers adaptive gamma selection, compression-aware token verification, and KV cache management for reduced latency. Use when: (1) Optimizing LLM serving throughput, (2) Implementing speculative decoding with draft models, (3) Managing KV cache for long-context generation, (4) Reducing inference latency for large language models. Triggers on: speculative decoding, KV cache, draft model, token verification, gamma selec...Votes: 0GitHub stars: 3
- Spike Driven Large Language Model SdllmSpike-driven Large Language Model (SDLLM) methodology. Eliminates dense matrix multiplications in LLMs through sparse addition operations using gamma-SQP two-step spike encoding. Reduces energy consumption by 7x while improving accuracy by 4.2% over previous spike-based LLMs.Votes: 0GitHub stars: 3
- Spike Driven Large Language ModelSpike-driven Large Language Model - Spike-based computation for large language models. Activation triggers: spike, driven, large, neuroscience, SNN.Votes: 0GitHub stars: 3
- Spike Nvpt Robust Visual PromptsSpike-NVPT: Noise-robust visual prompt tuning using bio-inspired temporal filtering and spike-based discretization. Parameter-efficient adaptation for pre-trained vision models with enhanced robustness to input perturbations. Keywords: Spike-NVPT, visual prompt tuning, bio-inspired, temporal filtering, noise robustness, spiking neural networks, computer vision.Votes: 0GitHub stars: 3
- Spoq Multi Agent Software EngineeringSPOQ (Specialist Orchestrated Queuing) - Multi-agent software engineering methodology with wave-based topological dispatch, dual validation gates, and Human-as-an-Agent integration for automated SE tasks. Activation: multi-agent orchestration, software engineering agent, agent coordination, SPOQ, task dispatch, validation gates, quality control, agent hierarchy, wave dispatch. Tags: multi-agent, software-engineering, orchestration, coordination, quality-control, task-dispatch, validation.Votes: 0GitHub stars: 3
- Staged Training Vlm Perception ReasoningStaged VLM post-training methodology - decomposing VLM capabilities into visual perception, visual reasoning, and textual reasoning stages with specialized data and RL-based perception learningVotes: 0GitHub stars: 3
- Stateful Streaming Transformer InferenceStateful streaming transformer inference methodology with persistent KV cache, Flash Queries prefetching, and multi-tenant continuous-batching for efficient streaming workloads.Votes: 0GitHub stars: 3
- Structured Search Llm ReasoningMethodology for improving LLM reasoning by making search tree structures explicit in reasoning traces. LinTree (arXiv:2605.31492) shows that adding parent pointers to linearized search traces significantly outperforms implicit reasoning and LLM-heuristic search. Use when optimizing chain-of-thought reasoning, implementing tree search in LLMs, designing reasoning agents, or analyzing search history conditioning in language models. Activation: structured reasoning, search tree LLM, linearized t...Votes: 0GitHub stars: 3
- Structured Sparse Attention EntityStructured-Sparse Attention methodology from arXiv:2605.22476 (May 2026). Blockwise resolvent-style attention operator achieving subquadratic sequence complexity O(n^(4/3)) for entity tracking by exploiting localized attention structure. Use when: efficient attention mechanisms, entity tracking, subquadratic transformers, sparse attention patterns, long-sequence reasoning.Votes: 0GitHub stars: 3
- Superintelligent Retrieval AgentSuperintelligent Retrieval Agent methodology for building retrieval-augmented systems that actively reason about information needs beyond black-box query issuance.Votes: 0GitHub stars: 3
- Symmetry Protected Lyapunov Equivariant RnnSymmetry-Protected Lyapunov Neutral Modes in Equivariant Recurrent Networks. Theoretical framework for when continuous attractors (zero Lyapunov exponents) are guaranteed by symmetry rather than tuning. Activation: equivariant rnn, symmetry-protected modes, Lyapunov neutral modes, continuous attractor, path integration, rotational memory.Votes: 0GitHub stars: 3
- System Dse DeepstackDeepStack methodology for design space exploration (DSE) in system-hardware co-design. Scalable and accurate performance modeling for distributed 3D-stacked AI systems. Use when performing early-stage system design optimization, hardware-software co-design, or DSE for AI accelerators. Keywords: DSE, design space exploration, system design, hardware co-design, distributed systems, AI accelerators.Votes: 0GitHub stars: 3
- Systems Engineering Threat ModelingAutomated threat modeling and security verification for cyber-physical systems (CPS) and AI-enabled systems. Use when: (1) Designing security architecture for CPS/IoT systems, (2) Performing automated threat modeling from system architecture models (SysML, DFDs), (3) Analyzing attack chains in LLM-enabled robotic or autonomous systems, (4) Mapping vulnerabilities to security controls (NIST 800-53, MITRE ATT&CK), (5) Verifying system safety properties (HyperLTL, reachability), (6) Building sec...Votes: 0GitHub stars: 3
- Temporal Ecological Network RobustnessTemporal structure analysis of ecological networks for understanding robustness and collapse mechanisms. Methods for modeling plant-pollinator networks with seasonal turnover, analyzing temporal dynamics, detecting bistable regimes, and predicting catastrophic transitions. Triggers: ecological network analysis, plant-pollinator dynamics, temporal network robustness, ecosystem collapse prediction, percolation analysis, bistable ecological systems, community resilience analysis.Votes: 0GitHub stars: 3
- Tensor Network Neurological PredictorTensor Network Feature Engineering methodology for multi-class neurological disorder prediction from MRI data. Uses tensor network decompositions to extract high-dimensional features from sparse medical imaging. Activation: tensor network MRI, neurological disorder prediction, tensor feature engineering, multi-class brain disorder, MRI tensor decomposition.Votes: 0GitHub stars: 3
- Test Time TrainingFramework for implementing Test-Time Training (TTT) - updating model weights during inference to adapt to continuous information streams. Use when designing adaptive AI systems, online learning for LLMs, breaking static train-deploy paradigm, or handling real-world tasks with evolving data.Votes: 0GitHub stars: 3
- Texture Misalignment Cnn PerceptionPerceptual misalignment of texture representations in convolutional neural networks — finds no connection between CNN Brain-Score and alignment with human texture perception, suggesting texture perception involves mechanisms distinct from object recognition CNNs. Based on arXiv:2604.01341.Votes: 0GitHub stars: 3
- Thermodynamic Networks ComputationThermodynamic Networks methodology for autonomous physics-based computation using non-equilibrium steady states. Identifies Negative Differential Conductance (NDC) as the critical property for computational expressivity. Applies to quantum dot networks, enzymatic reaction networks, and physical reservoir computing. Activation: thermodynamic networks, non-equilibrium computation, steady-state computing, NDC computation, physics-based computation, autonomous computation.Votes: 0GitHub stars: 3
- Training Free Looped Transformers"Training-free looped transformers — inference-time technique that loops a contiguous mid-stack block of layers in a frozen pretrained LLM to improve reasoning without fine-tuning. Use when: (1) improving a frozen LLM's reasoning at inference time, (2) retrofitting recurrence onto pretrained models without any training, (3) getting free accuracy gains on reasoning benchmarks with no additional training cost.Votes: 0GitHub stars: 3
- Transformer Guided Adaptive Diffusion AlzheimerMulti-Modal Graph Neural Network with Transformer-Guided Adaptive Diffusion for Preclinical Alzheimer Classification. Combines diffusion kernel (short-range) + multi-head attention (long-range) for brain network analysis. Activation: Alzheimer, preclinical AD, brain network, multi-modal GNN, diffusion kernel, transformer attention.Votes: 0GitHub stars: 3
- Transformer Prototype ReadoutUse prototype-based readout layers for transformer encoders to replace pooling methods (mean pooling, class token). Avoid information collapse with learned compression mechanism. Activation: prototype readout, transformer output layer, collapse-free attention, DDCL-Attention.Votes: 0GitHub stars: 3
- Transformer Warmstart Unit CommitmentMulti-Stage Warm-Start Deep Learning Framework for Unit Commitment. Transformer-based architecture for predicting generator commitment schedules with deterministic post-processing for physical feasibility, warm-start strategy for MILP solver, and confidence-based variable fixation. Use for power grid optimization, unit commitment problems, and MILP warm-starting with machine learning.Votes: 0GitHub stars: 3
- Trapped Ion Portfolio OptimizationEnd-to-end pipeline for large-scale portfolio selection with cardinality constraints using trapped-ion quantum computers. Use when: executing portfolio optimization on trapped-ion QPU hardware; solving QUBO subproblems via BF-DCQO; decomposing large portfolios via correlation-guided splitting; implementing two-stage post-processing for cardinality constraints; benchmarking quantum vs classical portfolio methods. Keywords: trapped-ion, portfolio optimization, QUBO decomposition, BF-DCQO, corre...Votes: 0GitHub stars: 3
- Tribe V2 Foundation ModelTRIBE v2 tri-modal foundation model methodology for in-silico neuroscience. Uses video/audio/language embeddings to predict whole-brain fMRI across 720 subjects.Votes: 0GitHub stars: 3