メニュー
このリポジトリの skills
SkillsMP は hiyenwong/ai_collection から 4,114 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
hiyenwong/ai_collection収集済み skill 4,114 件中 40 件を表示しています。
Brain digital twins execution semantics framework bridging computational modeling and neurobiological dynamics. Covers physically constrained executability, end-to-end workflow preservation, and neuromorphic implementation. Activation: brain digital twins,…
原文の言語: 英語
Framework for brain digital twins centered on execution semantics, bridging computational brain models to executable systems. Unifies fragmented data pipelines, model classes, and temporal scales. Activation: brain-digital-twins, execution-semantics,…
原文の言語: 英語
Brain digital twins execution semantics survey bridging computational neuroscience to neuromorphic systems - arXiv:2604.13574 (April 2026). Covers physically constrained executability taxonomy, execution regimes, hybrid-time correctness, and…
原文の言語: 英語
Brain-DiT universal multi-state fMRI foundation model with metadata-conditioned diffusion pretraining. Trigger words: Brain-DiT, fMRI foundation model, diffusion transformer, multi-state, metadata-conditioned
原文の言語: 英語
Brain-DiT universal multi-state fMRI foundation model methodology. Metadata-conditioned diffusion pretraining with DiT on 349,898 sessions across 24 datasets spanning resting, task, naturalistic, disease, and sleep states. Activation: brain-dit, fmri…
原文の言語: 英語
Brain-DiT v5 universal multi-state fMRI foundation model with pre-training and fine-tuning for zero-shot and few-shot brain decoding across multiple states. Supports cross-task, cross-subject, and cross-dataset fMRI analysis using diffusion transformer…
原文の言語: 英語
Brain-DiT v6 universal multi-state fMRI foundation model with metadata-conditioned pretraining across 24 datasets covering resting, task, naturalistic, disease, and sleep states. Use when working with fMRI foundation models, brain state decoding, or…
原文の言語: 英語
Brain-DiT universal multi-state fMRI foundation model methodology. Integrates diffusion transformer architecture with fMRI data for generative modeling and brain state analysis. Covers multi-state fMRI generation, brain state transition modeling, and…
原文の言語: 英語
Naturality Violation Score (NVS) for brain-DNN alignment beyond object-level comparison. Uses category theory to test whether brains and DNNs preserve the same transformations among stimuli. Covers approximate naturality, axis-resolved alignment analysis, and…
原文の言語: 英語
RE-CONFIRM framework for validating robustness of biomarkers discovered by brain foundation models from dynamic functional connectivity. Systematic evaluation of internal reliability, external reliability, and validity for clinical biomarkers. Activation:…
原文の言語: 英語
Analysis and mitigation of batch effects in fMRI foundation model embeddings. Activation: batch effects, fMRI foundation models, embedding quality.
原文の言語: 英語
Brain foundation model inversion methodology using Simulation-Based Inference (SBI) for stimulus reconstruction from synthetic neural activity. Enables reverse application of brain emulators like TRIBEv2 to recover stimuli or their properties from neural…
原文の言語: 英語
基于群体模板的脑图数据增强方法，用于改进单样本学习分类。使用连接脑模板（CBT）和图生成对抗网络（gGAN）从单一群体模板生成增强数据，提升阿尔茨海默病等疾病分类性能。触发词：脑图增强、数据增强、单样本学习、连接脑模板、CBT、图GAN、Alzheimer分类、brain graph augmentation、one-shot learning、CBT、gGAN。
原文の言語: 中国語
Graph Neural Network methods for brain connectivity analysis. Use when analyzing fMRI/EEG brain network data, modeling brain structure-function relationships, predicting cognitive outcomes from connectome data, or applying GNN to neuroscience problems.…
原文の言語: 英語
Brain-guided language model framework for robust reasoning - using task-fMRI signals from reasoning regions to enhance LLM performance across 10 models with up to 13% accuracy gain
原文の言語: 英語
脑网络高阶结构分析方法论。使用单纯复形和持续同调研究功能脑网络中的高阶交互（四节点以上），解释为何常规分析难以检测复杂高阶结构。触发词：高阶结构、持续同调、单纯复形、脑网络拓扑、higher-order interactions、persistent homology、simplicial complex。
原文の言語: 中国語
Brain-inspired attention mechanisms for neural networks - incorporating biological attention systems including thalamocortical circuits, pulvinar-mediated attention, basal forebrain modulation, and predictive processing. Implements biologically plausible…
原文の言語: 英語
Brain-Inspired Capture (BI-Cap) methodology for evidence-driven neuromimetic perceptual simulation. Models human perceptual processes for robust visual understanding. Activation: brain-inspired capture, neuromimetic perceptual, BI-Cap, evidence-driven…
原文の言語: 英語
Brain-Inspired Capture (BI-Cap) methodology for evidence-driven neuromimetic perceptual simulation in visual decoding. Trigger words: BI-Cap, neuromimetic, perceptual simulation, visual decoding, HVS
原文の言語: 英語
[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.]
原文の言語: 英語
Brain-inspired gating mechanism for Spiking Neural Networks that unlocks robust computation by incorporating dynamic conductance mechanisms. Addresses limitations of conventional LIF neurons that omit conductance dynamics inherent in biological neurons. Based…
原文の言語: 英語
类脑智能范式方法论。从神经科学视角重新思考智能形成与演化，提出Brain-like Neural Network (BNN)新范式，涵盖结构组织、学习机制、演化路径三大维度。触发词：类脑智能、BNN、神经科学智能、脑启发范式、brain-like neural network、intelligence paradigm、neuroscience AI。
原文の言語: 中国語
Brain-inspired Neural Cellular Automata (BraiNCA) for morphogenesis and motor control. Uses biological neighborhood structures beyond regular grids.
原文の言語: 英語
分析 Spiking Neural Networks (SNN) 和脑启发计算论文，提炼可复用的技术模式和实现指南。Use when analyzing papers about: spiking neural networks, brain-inspired computing, neuromorphic systems, biological learning rules, SNN architectures, or extracting implementation patterns from…
原文の言語: 複数言語
Brain-IT-VQA framework for visual question answering from fMRI brain signals. Decodes language tokens from brain activity and integrates with language model to answer visual questions. Use when: (1) Building VQA systems from brain signals, (2) Analyzing brain…
原文の言語: 英語
Brain-LLM alignment is driven by training-language dominance, not an inherent property of English. Tests with fMRI from 112 participants across English, Chinese, French and 7 LLMs (English-dominant, Chinese-dominant, multilingual). Baichuan2-7B reverses…
原文の言語: 英語
Brain-LLM analogy methodology for identifying grammar-specialized neurons in Large Language Models. Uses brain lesion study-inspired approaches to find POS-tag-specific neurons in Llama 3. Activation triggers: grammar neurons, LLM interpretability,…
原文の言語: 英語
脑MRI基础模型临床部署框架，基于FOMO25挑战赛发现。使用自监督学习(SSL)预训练大规模脑MRI基础模型，支持少样本学习和域外泛化，涵盖梗死分类、脑膜瘤分割和脑龄回归任务。Activation: brain MRI foundation model, FOMO25, clinical deployment, self-supervised learning, few-shot learning.
原文の言語: 中国語
结构脑网络可控性分析实用指南。基于网络控制理论计算平均可控性、模态可控性、最小控制能量等指标，研究脑网络状态转移能力。适用于脑网络分析、神经调控研究、认知控制研究。触发词：脑网络可控性、控制理论、结构连接、控制能量、平均可控性、模态可控性、controllability、control theory、structural brain network。
原文の言語: 複数言語
Brain-OF: First omnifunctional brain foundation model jointly pretrained on fMRI, EEG and MEG. Uses Any-Resolution Neural Signal Sampler, DINT attention with Sparse MoE, and Masked Temporal-Frequency Modeling for dual-domain pretraining. Pretrained on ~40…
原文の言語: 英語
Brain-OF unified foundation model for multiple neuroimaging modalities. Activation: omnifunctional model, multi-modal neuroimaging, unified brain analysis.
原文の言語: 英語
脑网络振荡同步计算框架 - 整合 Kuramoto相位动力学、时间延迟塑性、信息通量优化的统一方法论。用于研究脑网络同步性、频率选择、振荡耦合、信息传输效率。适用于网络神经科学、计算神经科学、脑动力学建模。
原文の言語: 中国語
Entropy-based active learning methodology for fair and efficient brain image segmentation. Use when: (1) performing brain MRI/CT segmentation with limited labeled data, (2) needing fair segmentation across demographic groups, (3) selecting most informative…
原文の言語: 英語
脑状态转移动力学与网络控制方法论。基于网络控制理论计算平均可控性、模态可控性、最小控制能量等指标，研究脑网络状态转移的能量需求与动态特性。Activation: brain state transition, network control, controllability, control energy, brain dynamics, structural connectome, functional connectome, 控制理论, 状态转移, 脑网络控制.
原文の言語: 中国語
Brain Stimulation Effects on Network Dynamics
原文の言語: 中国語
Source attribution framework for MEG-to-audio brain decoding that separates decoding performance into structural shortcuts, stimulus-evoked evidence, and contextual aggregation
原文の言語: 英語
Brain-to-speech synthesis from intracranial EEG using prosody feature engineering and transformer-based reconstruction — decodes speech with natural intonation and rhythm from neural signals. Maps neural activity to speech acoustic features including pitch…
原文の言語: 英語
Unified brain-to-text decoding framework for both speech production and perception in Mandarin Chinese. Uses shared neural representations across modalities with dual-decoder architecture. Activation: brain-to-text decoding, speech BCI, neural speech…
原文の言語: 英語
BraiNCA: brain-inspired neural cellular automata with complex brain-like topologies for morphogenesis and motor control. Long-range connections beyond Moore neighborhood. Triggers: neural cellular automata, morphogenesis, NCA, brain-inspired topology, motor…
原文の言語: 英語
BrainCast methodology for spatio-temporal forecasting of whole-brain fMRI time series. Uses dual-branch architecture with ST-CausalConv for spatial decoding and ST-Mixer for temporal prediction. Activation: fMRI forecasting, brain time series prediction,…
原文の言語: 英語