Menu
Skills in this repository
SkillsMP has collected 4,114 skills from hiyenwong/ai_collection. Open a skill to review its source and details.
hiyenwong/ai_collectionShowing 40 of 4,114 collected skills.
BrainDINO: Brain MRI Foundation Model for generalizable clinical representation learning. Self-distilled foundation model trained on 6.6M unlabeled axial slices from diverse brain MRI datasets, generalizing across heterogeneous endpoints without task-specific…
BrainFuse - unified infrastructure integrating realistic biological neural modeling and core AI methodology. Supports differentiable biophysical neuron simulation, 3000x GPU acceleration for ion-channel dynamics, and neuromorphic hardware deployment.
BrainJanus 统一脑模型方法论 - 整合脑、视觉和语言的首个统一框架，支持 any-to-any 生成（图像/文本到脑信号编码，脑信号到图像/文本解码）。使用 Unified Brain Tokenizer 和 All-in-One 自回归架构实现多模态脑信号处理。Activation: BrainJanus, unified brain model, brain encoding decoding, multimodal brain, 统一脑模型, 脑编码解码
Source text: Chinese
BrainJanus: first unified brain model integrating brain, vision, and language via autoregressive Transformer. Unified Brain Tokenizer quantizes neural dynamics into discrete tokens aligned with visual/linguistic representations. Supports any-to-any generation…
BrainPICM 个性化社区掩码方法论 - 基于渐进式个体化社区感知掩码的脑网络自监督学习框架。使用最优传输进行 ROI 到社区的软分配，课程式掩码策略逐步引入低置信度区域，偏差感知聚合模块量化功能重组。适用于 fMRI 脑网络分析和神经精神疾病诊断。Activation: BrainPICM, personalized community, progressive masking, brain network SSL, optimal transport brain, 个性化社区, 渐进式掩码, 脑网络自监督
Source text: Chinese
Lightweight self-supervised representation learning for fMRI using positive-only data pairs, achieving strong cross-task generalization without large-scale pretraining
Branched optimal transport framework for inferring stimulus-to-reaction propagation architectures in brain networks. Uses anisotropic branched OT where concavity of flux cost promotes aggregation and branching. Activation: branched optimal transport, stimulus…
BRIDGE神经多样性检测方法论。整合图规范建模和深度生成网络，创建神经典型发育轨迹参考，评估个体神经偏离程度。适用于神经发育障碍诊断、脑龄预测、连接组分析。触发词：神经多样性、神经发育障碍、脑龄、规范建模、neurodivergence、brain age。
Source text: Chinese
Position and survey paper identifying convergence of neuroscience, AGI, and neuromorphic computing. Synaptic plasticity, spike-based communication, and multimodal association as design principles for next-gen AGI. arXiv:2507.10722
Calibrated measurement framework using the Brody exponent β as a quantitative measure of short-range exclusion in 2D spatial point processes. Originally from quantum chaos level-spacing statistics, now calibrated for spatial analysis with corrected CSR…
BSViT: Burst Spiking Vision Transformer with Dual-Channel Burst Spiking Self-Attention for expressive and efficient visual representation learning. Addresses binary spike capacity limits with burst encoding and local attention masking. Activation: bsvit,…
Excitation-driven data generation framework for building thermal dynamics modeling and control. BuilDyn enables customizable excitation strategies for control-oriented data generation, supports sampling from representative building distributions, and provides…
并行训练时间精确脉冲神经网络的方法论。使用并行关联扫描实现44倍加速，机器精度脉冲时间求解器避免时间离散化近似，支持端到端事件驱动SNN训练。Activation: spiking neural network, snn training, parallel scan, spike timing, event-based, neuromorphic, 并行训练, 脉冲时间.
BUS: Brain-Inspired Unsupervised Self-Reflection for VLM reasoning. Uses backward prediction (neuroscience concept) to enable label-free self-reflective reasoning in Vision-Language Models. Compatible with SFT and RL, tested on 8 benchmarks. Activation: BUS,…
CaMBRAIN - 首个因果 Mamba 状态空间模型用于 EEG 信号的实时连续推理。解决传统注意力模型的二次复杂度问题，通过多阶段自监督训练实现长程记忆保持和线性时间复杂度。
Source text: Chinese
CaMBRAIN methodology for real-time continuous EEG inference using causal Mamba state space models. First model enabling long-range streaming inference of variable-length EEG signals with >10x higher throughput.
Canonical Functionalism — a mathematical refinement of computational functionalism about consciousness that identifies consciousness-relevant functional organization with a system's canonical functional structure: the minimal state-transition structure…
Convolutional Neural Network framework for detecting gaseous microemboli (GME) during cardiac procedures using transthoracic ultrasound. Activation triggers: emboli detection, cardiac ultrasound, microemboli GME, surgical safety, transcatheter monitoring
DIRECT mechanism for causal learning with neural assemblies - local plasticity-based directional learning without backpropagation. Enables neural assembly networks to internalize causal directionality through projection, local plasticity control, and sparse…
Causal state intervention framework for controlling human behavioral outcomes through targeted latent state manipulation. Demonstrates that within-person variability belongs to dynamic latent state, and human outcomes are controllable through interventions…
颅内脑电图因果脑网络研究方法论。使用 CCEP（皮质-皮质诱发电位） 直接表征脑区间功能连接和方向性。 触发词：因果脑网络、iEEG、CCEP、皮质诱发电位、方向性连接、 causal brain network, intracranial EEG, cortico-cortical evoked potential。
Source text: Mixed languages
Counter-Diabatic QAOA (CD-QAOA) methodology for peptide structure prediction on tetrahedral lattices. Accelerates convergence via counter-diabatic driving terms, validated against HF/DFT/MD/H-REMD. Use when: (1) quantum optimization for molecular/peptide…
Counterfactual Policy Optimization for enforcing causal consistency between visual perception and textual reasoning in LVLMs. Cross-modal counterfactual enhancement regularizes policy via discrepancy maximization.
CFSPMNet - Cross-subject Fourier-guided Spatial-Patch Mamba Network for EEG Motor Imagery Decoding in Stroke Patients. Use when working with MI-EEG decoding, cross-subject BCI for stroke rehabilitation, Mamba-based EEG models, or Fourier-domain token…
无突触可塑性的混沌冻结方法论。引入Onsager反应项使神经网络动力学稳定化，不依赖Hebbian学习即可抑制混沌波动。适用于RNN稳定性分析、神经动力学控制。触发词：混沌冻结、Onsager反应、神经网络稳定性、混沌抑制、RNN动力学、chaos freezing、Onsager reaction、gradient dynamics。
Source text: Chinese
Mechanistic reasoning framework for distillation in large output spaces - two-phase process of shortlisting candidates followed by fine-grained reasoning, consistently outperforming standard distillation.
CHASMBrain层级化Mamba架构用于图像到fMRI编码。双流Mamba设计分离全局语义token和局部空间patch处理，粗到细策略实现ROI级到voxel级预测，在NSD数据集上Pearson相关达0.429。
Source text: Chinese
CheeseBench benchmark for evaluating LLMs on classical rodent behavioral neuroscience paradigms. Includes 9 tasks covering water maze, T-maze, Morris water maze, and other established behavioral tests. Cross-paradigm evaluation for neuroscience AI systems.
CHIA framework for principled agentic AI-driven hardware/software co-design. Treat hardware/software design flows as directed cyclic graphs (CHIA loops) with nodes executing SoC design tools, simulators, AI models, and evolutionary agents. Supports Chipyard,…
Conductance-based neuronal network model for chloride-dependent seizure dynamics. Models how activity-dependent chloride dynamics drive seizure evolution and stage transitions. Activation: computational-neuroscience, seizure, chloride, conductance-based,…
Expert guidance for Chrome extension development with Manifest V3. Use when building Chrome extensions, browser extensions, or working with Chrome APIs. Triggers on: chrome extension, browser extension, manifest v3, chrome api, extension development.
Computational modeling methodology for chronic stress as excitatory-inhibitory (E/I) balance perturbation in recurrent working-memory networks. Use when modeling stress effects on prefrontal cortex, studying resilience mechanisms, or analyzing E/I balance…
Circuit-Level Noise Estimation via Shuttling in Plaquette Circuits. Use when analyzing quantum algorithms, complexity bounds, quantum ML architectures, or quantum error correction involving mathematical analysis and statistical methods.
Circuit-level spiking neuron model for hardware robustness analysis. Studies how transistor-level variations affect SNN reliability on neuromorphic chips. Activation: circuit-level SNN, neuromorphic hardware reliability, transistor variation spiking, hardware…
Circulate-Firing Spiking Neural Network Training methodology - advancing direct training with CFSN model, TSL-SG learnable gradients, and PNB-Loss. Addresses SNN training bottlenecks: limited information capacity and imprecise gradient propagation.…
Citizen science and video games framework for cognitive science research. Transforms players into research participants through gamified experiments, enabling large-scale data collection for understanding human cognition. Activation: citizen science, video…
CITRAS-FM tiny 7M-parameter time series foundation model with covariate-informed zero-shot forecasting using Shifted Attention and CovSynth synthetic covariate generation
Clustered Label-Agnostic Federated Learning (CLAD) framework for anomaly detection in distributed systems. Combines unsupervised clustering with supervised detection via DM²A (Dual-Mode Multi-Stage Aggregation), enabling privacy-preserving anomaly detection…
Classical probability model that reproduces the disjunction effect in human decision making without violating the law of total probability. Use when analyzing the disjunction effect, Prisoner's Dilemma decision paradox, quantum-like cognition models,…
Classical Shadow Estimation of Unitary Channels (CSEU) methodology for efficient quantum process learning with Heisenberg-limited query complexity.