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.
AUTOPILOT VQA: Benchmarking Vision-Language Models for Incident-Centric Dashcam Understanding. Recent advances in Vision-Language Models, Large Language Models, and Multimodal Large Language Models have improved autonomous driving tasks such as scene…
DexVerse: A Modular Benchmark for Multi-Task, Multi-Embodiment Dexterous Manipulation. Building general-purpose dexterous manipulation policies requires benchmarks that go beyond isolated tasks to systematically evaluate policies across diverse interaction…
SpikeProphecy: First large-scale benchmark for causal, autoregressive neural population spike-count forecasting. Introduces population metric decomposition (temporal fidelity, spatial pattern accuracy, magnitude-invariant alignment) on 105 Neuropixels…
Methodology for trading inference-time compute to improve adversarial robustness in LLMs through repeated sampling and output filtering.
JGRA framework for assessing robustness in NISQ noise-aware Quantum Neural Networks via Jacobian geometry. Captures model sensitivity to parameter perturbations induced by noise through entropy-matched noise calibration, noise-aware training, and…
Space-efficient streaming attention approximation using tight bounds for KV cache compression in transformer architectures.
Zero-shot imagined speech decoding from MEG via imagined-to-listened cross-condition mapping. Trains models to map imagined MEG responses to listened responses, then decodes using listened-only decoder. Three-stage pipeline: (1) mapping imagined→listened MEG,…
Reinforcement learning framework for neural model editing where agents learn to modify models via reward feedback instead of manually engineered algorithms
Analog Quantum Asynchronous Event-Based Graph Neural Network (QA-AEGNN) — implementing event-based GNNs on neutral-atom quantum processors via Rydberg Hamiltonian programming. Maps streaming event data to trapped atom arrays where geometric proximity reflects…
BBGKY-ISM quantum error mitigation using Feynman's clock Hamiltonian with polynomial overhead (arXiv: 2607.06752)
Quantum randomness certification framework using measurement incompatibility witnesses — bounds classical eavesdropper capabilities via semi-definite programming using generalised robustness as a geometric incompatibility measure. Use when certifying quantum…
Operator Kirigami methodology for symmetry conservation in quantum algorithms — cut-and-fold technique for preserving non-Abelian symmetries in Trotterized quantum circuits by orthogonal projection and unitary rotation folding.
Parameter-efficient Quantum Multi-task Learning (QMTL) methodology. Replaces conventional task-specific linear heads with fully quantum prediction heads in hybrid architectures. Quantum head parameters scale linearly with task count vs quadratic for classical…
Quantum error detecting codes using Pauli group variance geometry — going beyond stabilizer codes via algebraic structure of Pauli operators. Covers variance-based code construction, Pauli weight distribution analysis, higher-than-stabilizer detection rates,…
Penalty-free quantum annealer pipeline for portfolio optimization. Removes cardinality penalty from QUBO to reduce chain-break fractions from 92% to 0.04%, enforces feasibility classically post-sampling. Based on arXiv:2605.17628.
Bottom-up engineering methodology for non-Abelian topological order using Floquet-engineered synthetic magnetic fields and Bayesian-optimized adiabatic state preparation.
Phase model analysis methodology for M-current effects on neural synchronization in hippocampal networks - theoretical framework for understanding neuromodulatory control of synchrony
Phase-reference control methodology for generating steady-state entanglement in open quantum systems using phase-sensitive reservoirs and local dissipation. Activation: steady-state entanglement, phase-sensitive reservoir, open quantum systems, Gaussian…
Algorithmic advantage of gate-based photonic quantum neural networks over classical ANNs. Use when comparing QNN vs ANN performance, evaluating quantum neural network expressivity via effective dimension, designing photonic quantum classifiers, or analyzing…
Physics-aligned real-to-sim-to-real data engine for deformable object manipulation. SIM1 grounds simulation in physical world through metric-consistent digital twins, elastic dynamics calibration, and diffusion-based trajectory generation. Use for: deformable…
Source text: Chinese
Physics-guided and physics-informed quantum machine learning methodologies — embedding physical priors (symmetries, conservation laws, Hamiltonians) into VQCs, QNNs, and quantum classifiers for improved accuracy, trainability, and physical consistency
Physics-informed LLM framework (VF-QCTRL) for general quantum control synthesis combining symbolic reasoning with optimization. Proposes analytic control ansätze and refines parameters through feedback loops. Use when designing LLM-driven quantum control,…
Neuro-fuzzy framework for quantum error attribution using physics-informed machine learning. Combines ANFIS with physics-grounded features (Bhattacharyya Veto, Data Processing Inequality) to distinguish software bugs from hardware noise in quantum processors.…
Use Physics-Informed Neural Networks (PINNs) for quantum pulse optimization and noise-aware gate fidelity. Specifically for optimizing quantum control pulses in exchange-only spin qubit systems, handling charge noise, and maximizing gate-level fidelity…
Scalable distributed quantum computing architecture using photonic integration of designed molecular quantum nodes. Combines solid-state spin defects in diamond (NV/SiV centers) with nanophotonic waveguide networks for entanglement distribution. Applies…
Hardware-aware design platform for fault-tolerant quantum computers (FTQCs). Computes logical performance from device physics using Kraus operators, Hamiltonian-Lindblad dynamics, and quantum channels across four sampler classes.
Point-group symmetry analysis of many-electron wavefunctions on quantum computers. Ancilla-free hybrid method for abelian and non-abelian groups using orbital rotations from representation matrix eigenvectors, tensor-network encoding, and error mitigation for…
Post-quantum cryptographic analysis of network protocol stacks and message transformations. Evaluates security of cryptographic operations across multiple protocol layers against quantum attacks. Use when: (1) analyzing protocol stack security, (2) evaluating…
Post-Quantum Cryptography (PQC) migration framework for IoT-based healthcare systems. Provides systematic approach for transitioning healthcare infrastructure from classical to post-quantum cryptographic standards (NIST ML-KEM, ML-DSA) while maintaining…
NIST-standard PQC migration (ML-KEM + ML-DSA) for pharmacovigilance and healthcare data systems. Use when designing post-quantum security for adverse event reporting, clinical observation systems, or any healthcare pipeline handling sensitive patient data…
A型钾电流介导的神经元增益控制机制。研究IA作为减法抑制与除法抑制之间的开关，通过动力学系统分析理解神经元如何自调节抑制效果。适用于计算神经科学、神经元建模、增益控制研究。触发词：A型钾电流、增益控制、抑制模式、IA电流、神经元增益、divisive inhibition、subtractive inhibition、gain control、potassium current。
Source text: Mixed languages
Scenario-based optimization framework for predictive maintenance scheduling under uncertainty. Integrates calendar-based, usage-based, and condition-monitoring (RUL) information into unified finite-horizon decision framework. Use when: (1) optimizing…
Cohomological structure analysis of prime numbers using iterative maps, linking prime irregularities to physical systems including statistical mechanics and quantum mechanics.
Cohomological structure analysis methodology for prime numbers — iterative maps predicting prime growth, cohomological equation solutions, and connections between statistical mechanics, quantum mechanics, and number theory. The logarithmic integral function…
PRISM (Probabilistic Recurrent Intention Switching Model) methodology for multi-intention inverse reinforcement learning. Uses lightweight recurrent networks for intention switching with closed-form EM solution. Activation: 多意图 IRL, intention switching,…
Progressive Swapping to the Middle (PSM) protocol methodology for efficient entanglement distribution in quantum networks with imperfect quantum memories. Combines nested entanglement swapping with memory-aware scheduling. Use when designing quantum network…
前瞻编码与路径整合的自组织神经网络框架。揭示连续吸引子网络(CANNs)如何通过赫布塑性、发放率适应和全局抑制自组织形成，实现前瞻性编码和路径整合功能。
精神病早期阶段脑动力学临界性scaling偏差研究方法论。结合重整化群(RG)框架与多种scaling分析方法，揭示临界 regime内的动力学重组而非临界性丧失。
Source text: Chinese
Q-BIOLAT: Binary latent protein fitness landscapes for quantum annealing optimization. Maps protein sequences to binary latent spaces via pretrained protein language models, then uses quantum annealing (D-Wave) for fitness landscape exploration and protein…
Q-PhotoNAS: Hybrid Quantum Neural Architecture Search framework for photonic quantum-classical models using genetic algorithm and learnable phase encoding