메뉴
이 저장소의 skills
SkillsMP는 hiyenwong/ai_collection에서 4,114개의 skill을 수집했습니다. skill을 열어 소스와 세부 정보를 확인하세요.
hiyenwong/ai_collection수집된 skill 4,114개 중 40개를 표시합니다.
LLM as Clinical Graph Structure Refiner for EEG seizure diagnosis. Two-stage framework using LLMs to refine graph edges for cleaner, more interpretable graph representations in automated seizure detection. Accepted by IJCAI-ECAI 2026.
원문 언어: 영어
LLM自动化标注情绪轨迹fMRI解码方法论。使用多目标回归框架、动态功能连接(DFC)、图论可解释AI解码连续情绪维度。
원문 언어: 중국어
LLM与人类神经语义表征收敛性研究方法论。使用伪超扫描MEG实验设计、维度分解的跨脑编码建模、十维语义空间评估，揭示LLM选择性对齐人类共享神经语义的维度依赖特性。
원문 언어: 중국어
LLM-assisted post-quantum cryptography coding security patterns — analyzing secure coding drift, constant-time execution requirements, side-channel resistance, and gamified remediation strategies for PQC implementations.
원문 언어: 영어
Framework for evaluating and training LLMs to assist in migrating pre-quantum cryptographic code to post-quantum counterparts. Systematic assessment methodology measuring code correctness, security preservation, and functional equivalence during PQC…
원문 언어: 영어
Large Language Models in Process Systems Engineering (PSE) - systematic survey of LLM applications across seven categories with capability assessment
원문 언어: 영어
LLM-Human neural semantic convergence methodology - dimension-resolved interbrain encoding modeling comparing LLM-derived and human-shared neural semantic representations across 10 semantic dimensions. Use when: LLM brain alignment, semantic representation…
원문 언어: 영어
LLM睡眠-记忆巩固机制：借鉴生物睡眠的记忆整理原理，实现语言模型的自我修改与记忆整合。核心概念：睡眠阶段用于记忆重放、权重优化、灾难性遗忘缓解。触发词：LLM睡眠、记忆巩固、自我修改、sleep paradigm、记忆整理、遗忘缓解。
원문 언어: 중국어
Structured Concept Evolution (SCE) — search framework pairing LLMs with structured algebraic mutation grammars to discover quantum LDPC code families. Evolves structured concepts (algebraic specifications + executable programs) via hierarchical mutations on…
원문 언어: 영어
Identifying structural design principles (local cycles) that shape computational abilities of recurrent neural networks. Found that 2- and 3-cycles strongly enhance computational power, and biologically-inspired interneurons dramatically increase capacity.
원문 언어: 영어
Local Pheromone Network methodology for sparse, local, manually updated neural networks without backpropagation. Uses pheromone-weighted Hebbian updates with short-term/long-term synaptic traces, consolidation, and replay. Achieves partitioned memory…
원문 언어: 영어
本地基座模型强化学习对齐工程实践 - 涵盖 RLHF/DPO/GRPO 算法选型、显存优化、框架选择、数据工程与全流程实施指南
원문 언어: 중국어
LOCO (Low-rank Cluster Orthogonal) weight modification for backpropagation-free SNN training. Perturbation-based non-BP learning with O(1) parallel time complexity, enabling deep SNN training (10+ layers) with continual learning capability. Activation:…
원문 언어: 영어
LogQ algorithm reformulated as classical non-linear continuous relaxation for QUBO problems. Use when: solving portfolio optimization, fleet optimization, charging station placement, or any QUBO combinatorial problem; implementing quantum-inspired classical…
원문 언어: 영어
Loop composition methodology for quantum algorithms. Models program control flow (branching + looping) in quantum circuits using quantum walk formalism. Addresses limitations of straight-line quantum circuit model for variable-length subroutines in…
원문 언어: 영어
Loss-biased fault-tolerant quantum error correction methodology using fast autoionization in alkaline-earth atoms. Implements practical fault-tolerant quantum computing with sub-millisecond QEC cycles and high encoding efficiency. Use when: (1) Analyzing…
원문 언어: 영어
Lottery BP methodology for scalable quantum error correction decoding. Introduces randomness during belief propagation decoding to improve accuracy by 2-8 orders of magnitude for topological codes (surface, toric, BB codes). Use when: quantum error…
원문 언어: 영어
Low-frequency (alpha-band) activity shapes fine-scale information routing in early visual cortex — alpha oscillations in V1 carry spatially specific figure-ground information and modulate inter-areal V1-V4 coupling during visual processing, supporting the…
원문 언어: 영어
Mathematical framework for compositional computation in inhibition-dominated threshold-linear networks via low-rank gluing rules. Proves how structural modularity enables functional compositionality - component subnetworks' fixed points determine global…
원문 언어: 영어
Behavioral and brain alignment methodology between Large Reasoning Models (LRMs) and human game learners, using fMRI-validated complex gameplay datasets. Activation: LRM brain alignment, reasoning model cognitive neuroscience, AI human game learning, frontier…
원문 언어: 영어
LSFormer: Local Structure-Aware Spiking Transformer. Replaces global self-attention with dilated local windows and spiking response pooling for energy-efficient SNNs. Keywords: spiking transformer, local attention, SNN, energy-efficient, LSFormer, spiking…
원문 언어: 영어
Multi-Atlas Disentangled Connectivity LEarning (MADCLE) methodology for brain disorder identification from functional connectivity (FC) matrices. Addresses atlas dependency heterogeneity by jointly encoding FC matrices from different brain atlases with…
원문 언어: 영어
Magic-Informed Quantum Architecture Search (QAS) methodology using Monte Carlo Tree Search with Graph Neural Networks for quantum circuit design. Use when designing quantum circuits with controlled nonstabilizerness (magic) levels, when optimizing quantum…
원문 언어: 영어
Multi-Scale Adaptive Graph Network (MAGNet) for learning structural-functional brain representations. Models structure-function coupling for cognitive insight.
원문 언어: 영어
Methodology from Anthropic research (Jun 2026) on benchmarking LLM capability for chemistry tasks, specifically NMR spectral analysis and molecular structure elucidation. Opus 4.7 achieves competitive accuracy with ChemDraw/MestReNova on hydrogen NMR (±0.079…
원문 언어: 영어
Mamba forecaster methodology for implicit behavioral decoding from next-step spike forecasts at population scale. A single sequence model trained only on next-step Poisson rate prediction produces predicted firing rates that decode animal behavior better than…
원문 언어: 영어
Implicit Behavioral Decoding from Next-Step Spike Forecasts at Population Scale — using a single Mamba state-space model trained only on next-step spike counts (Neuropixels scale) to simultaneously forecast neural population activity and decode behavioral…
원문 언어: 영어
Implicit Behavioral Decoding from Next-Step Spike Forecasts at Population Scale — Mamba forecaster methodology for closed-loop BCI. A single Mamba model trained on next-step spike counts at Neuropixels scale simultaneously predicts future neural activity and…
원문 언어: 영어
Mamba-based spike forecaster methodology for closed-loop BCI. A single Mamba model trained on next-step spike counts at Neuropixels scale simultaneously predicts neural activity and decodes behavioral state, outperforming linear decoders on raw spikes. arXiv:…
원문 언어: 영어
Many-body chirality methodology for topological stabilizer states — formulated as obstruction to complex conjugation via finite-depth local operations, with four-partite obstruction and intrinsic imaginarity.
원문 언어: 영어
Many-body super- and subradiance in ordered atomic arrays. Studies collective light-matter interactions in subwavelength-spaced atom arrays with programmable photon-mediated interactions. Activation: superradiance, subradiance, ordered atomic arrays,…
원문 언어: 영어
Mass conservation as inductive bias for self-organized criticality in neural cellular automata reservoirs. Demonstrates 1.27× faster evolution with comparable downstream performance. Activation: self-organized criticality, neural cellular automata, reservoir…
원문 언어: 영어
MAST (Multi-channel pseudo-event SNN with Adaptive Spiking Temporal integrators) — first SNN-based detector for AI-generated videos. Converts inter-frame residuals into pseudo-events processed by spike-driven temporal branch with learnable per-channel time…
원문 언어: 영어
Maximum entropy principle for neural network connectivity — normative framework for understanding how task constraints shape neural connectivity structure without gradient descent.
원문 언어: 영어
Multi-Band Brain Net (MBBN) — Transformer-based framework for frequency-specific spatiotemporal brain dynamics from fMRI. Integrates biologically-grounded frequency decomposition with multi-band self-attention for cognitive and psychiatric applications. Use…
원문 언어: 중국어
麦肯锡工作法39个工作习惯——结构化问题解决、高效沟通、时间管理与持续成长的方法论
원문 언어: 중국어
MCTS-based quantum data encoding discovery methodology. Use Monte Carlo Tree Search to discover optimal data encoding circuits for quantum-classical neural networks. Evaluates encoding strategies by effective rank correlation rather than entanglement…
원문 언어: 영어
Dynamical mean-field theory for low-rank recurrent networks with firing-rate adaptation. Identifies four oscillatory regimes and bifurcation mechanisms linking chaos, Hopf bifurcation, and noise-sustained oscillations to biological rhythms (Up-Down states,…
원문 언어: 영어
Dynamical mean-field theory for random recurrent networks with low-rank structure and firing-rate-driven adaptation. Identifies four oscillatory regimes: static coherent, noise-sustained oscillations, stochastic switching, global limit cycle. Explains…
원문 언어: 영어
Measurement-based soft PCA framework using entropy-regularized Fermi-Dirac filters for quantum principal component analysis without eigenvector recovery. Enables dimension-independent sample complexity O(1/eta^2) for fractional-rank scoring. Use when: quantum…
원문 언어: 영어