Menu
Skills dans ce dépôt
SkillsMP a collecté 4 114 skills depuis hiyenwong/ai_collection. Ouvrez un skill pour examiner sa source et ses détails.
hiyenwong/ai_collectionAffichage de 40 skills collectés sur 4 114.
Von Economo神经元快速通道假说 - 生物速度-准确性权衡的计算模型。VENs作为快速稀疏投射通路，在复杂社会认知中实现快速决策。首次建立VENs的计算模型，解释其在快速社会决策中的功能。Activation: Von Economo neurons, VEN, speed-accuracy tradeoff, fast lane hypothesis, social cognition, spiking neural network, biological decision making.
Langue du texte source : anglais
Communicative Language Symbolism Routing (CLSR): a test-time framework where multiple LLM agents autonomously invent, evolve, and share compact Language Symbolism Frameworks (LSFs) for efficient multi-agent reasoning. A latent-free router adaptively selects…
Langue du texte source : anglais
PDEFlow: an autonomous agentic framework that turns user-level ODE/PDE descriptions into solver-backed neural-operator pipelines. Links problem specification, data generation, operator training, and checkpoint-based inference via a stateful input graph and…
Langue du texte source : anglais
Uncertainty-aware neuro-symbolic multi-agent framework for adaptive ransomware detection. Fuses semantic representation evidence with behavioural forensic telemetry, using Monte Carlo Dropout for epistemic uncertainty quantification. A risk-uncertainty…
Langue du texte source : anglais
BCPNN (Bayesian Confidence Propagation Neural Network) native explainability framework. First XAI taxonomy for BCPNN, mapping architectural primitives to attribution, prototype, concept, counterfactual, and mechanistic explanations. Introduces 16…
Langue du texte source : anglais
Large-scale AI benchmarking methodology for cancer detection models. Evaluates tumor-detection AI across tumor size, location, demographic subgroups, and imaging protocols using 85,355 CT scans and 12 models. Use when: benchmarking medical AI models,…
Langue du texte source : anglais
Continuous Multi-Mode Scheduling（CMMS）基准测试平台用于边缘集群调度算法公平比较。统一控制器接口、闭环负载驱动、双指标SLO评分（原始SLO vs 稳态SLO），揭示控制器排名的配置依赖性和切换成本。Activation: edge cluster scheduling, heterogeneous scheduling, SLO benchmark, CMMS, RL scheduling, adaptive benchmark, edge-cloud continuum.
Langue du texte source : chinois
AI-powered vulnerability discovery methodology from Anthropic's Project Glasswing - using frontier models (Mythos Preview) to find critical security vulnerabilities in open-source and enterprise software.
Langue du texte source : anglais
GNN-based drug toxicity prediction explainability methodology with Gap Taxonomy (GAP-1 to GAP-4) for systematic analysis of explainability limitations. Uses GNNExplainer on MPNN models trained on Tox21 benchmark.
Langue du texte source : anglais
Maximum-Caliber Deviation framework bridging Integrated Information Theory (IIT) with the Free Energy Principle (FEP). Defines information as deviation from constrained maximum-caliber path ensembles, re-derives IIT cause/effect repertoires from variational…
Langue du texte source : anglais
Novel framework for cross-subject neural alignment without shared stimuli, using pretrained ANN as common scaffold to enable generalizable decoders and cross-subject prediction
Langue du texte source : chinois
OBLIQ-Bench methodology for exposing overlooked bottlenecks in modern retrievers with latent and implicit queries. Identifies oblique queries seeking documents that instantiate latent patterns. Reveals retrieval-verification asymmetry where LLMs recognize…
Langue du texte source : anglais
Methodology from Anthropic research (Apr 24, 2026) — AI agent marketplace experiment where Claude acts as negotiator buying/selling on behalf of employees, testing agentic negotiation and decision-making capabilities.
Langue du texte source : anglais
Anthropic research (Jun 18, 2026) — Project Fetch phase two results on AI agent capability for offensive cyber operations and exploit development assessment framework.
Langue du texte source : anglais
Methodology from Anthropic's Project Glasswing — using frontier AI models for large-scale cybersecurity vulnerability discovery and remediation. Based on May 22, 2026 initial update.
Langue du texte source : anglais
Anthropic's autonomous AI shopkeeper experiment investigating real-world business task performance, multi-agent coordination (CEO + worker), and emergent behaviors in commercial settings.
Langue du texte source : anglais
Accelerating benchmarking of functional connectivity (FC) modeling via structure-aware coreset selection for large-scale fMRI datasets. Reduces combinatorial explosion in model-data evaluation pairs. Activation: functional connectivity, fMRI benchmarking,…
Langue du texte source : anglais
Build and analyze knowledge graphs from research literature. Automated pipeline: arxiv search → entity extraction → KG construction → vector embeddings → semantic search → skill pattern extraction. Use when user asks to analyze papers, build research…
Langue du texte source : anglais
Graded Input-based Quantization Hierarchy for efficient LLM generation. Dynamic precision assignment based on activation magnitudes as computational importance proxy. Unifies quantization and sparsification.
Langue du texte source : anglais
Piper framework for user-controllable distributed training that decouples parallelism strategy from runtime implementation using unified global training DAG intermediate representation. Use for distributed ML training, parallelism strategy design, and…
Langue du texte source : anglais
Surviving by Serving (SBS) principle - functional relevance drives self-organization in complex adaptive systems with multi-agent resource transformation
Langue du texte source : anglais
Market predictability framework distinguishing epistemic uncertainty (reducible) from aleatoric uncertainty (irreducible) in financial markets. Based on the thesis that markets are not random but hard to predict — with profound implications for investment…
Langue du texte source : anglais
Forward gradient estimation methodology for training parameterised quantum circuits (PQCs). Introduces QUIVER adaptive optimiser that recovers SPSA, random coordinate descent, and parameter-shift rule as limiting cases. Enables efficient training of 60-qubit…
Langue du texte source : anglais
Analog Interaction Systems (AIS) methodology for energy-efficient generative modeling on neuromorphic hardware. Bridges gap between software-defined generative models and fixed physics-determined differential equations in analog circuits. Use when designing…
Langue du texte source : anglais
Democratic Inverse Constitutional AI (ICAI) methodology that derives steering principles from human preferences through structured debate, capturing the reasoning underlying human judgments rather than just final choices.
Langue du texte source : anglais
Group intervention-based causal discovery for identifying causal structure in deep neural network subsystems — extending causal discovery from single neurons to functional subnetwork groups. Activation: causal discovery, deep network, subsystem, group…
Langue du texte source : anglais
Hebbian Fast-Weight (HFW) modules integrated into Vision Transformer architectures for few-shot learning. Activation triggers: hebbian fast weights, hebbian ViT, fast synaptic updates, transformer meta-learning, few-shot transformer, hebbian plasticity…
Langue du texte source : anglais
HES (High-Entropy Sum) methodology from arXiv:2605.22389 (May 2026). Training-free metric for LLM reasoning data selection: sums entropy of top-k highest-entropy tokens per reasoning sample. Effective across SFT, RFT, and RL training paradigms. Use when: LLM…
Langue du texte source : anglais
Methodology for high-precision numerical evaluation of multivariate hypergeometric functions using Pfaffian systems and contour restriction. Applicable to quantum field theory, string theory, number theory, and statistics computations.
Langue du texte source : anglais
ICE review workflow for consolidating knowledge from completed tasks into reusable memory and future leverage.
Langue du texte source : anglais
Fault-tolerant error detection using the Iceberg [[2m, 2m-2, 2]] quantum error-detecting code. Implements beyond-break-even error detection for multi-qubit gates on trapped-ion quantum computers. Keywords: quantum error detection, Iceberg code,…
Langue du texte source : anglais
Intrinsic Computational Functionalism methodology — From Observer-Relative Maps to Observer-Independent Structures. Addresses observer-relativity problem in computational theories of consciousness with operationalizable criteria.
Langue du texte source : chinois
Jeffreys Flow framework for robust Boltzmann generators and rare event sampling. Addresses mode collapse in multi-modal distributions using Jeffreys divergence + Parallel Tempering distillation. Use when: sampling rough energy landscapes, Boltzmann…
Langue du texte source : anglais
Methodology for analyzing analytic properties of Jost functions in quantum scattering theory via parameter-dependent ODEs (Poincare-Picard theorem). Applies to scattering matrix analytic continuation, complex energy plane analysis, and quantum scattering…
Langue du texte source : anglais
KL Agreement Trap Termination (KAT) for on-policy distillation. Detects persistent low-KL agreement traps and terminates early to improve training efficiency.
Langue du texte source : anglais
Bridging Krylov complexity theory with universal analog quantum simulation — using Lanczos algorithm and Krylov subspace growth to characterize computational power of analog quantum simulators. Activation: Krylov complexity, analog quantum simulator, Lanczos…
Langue du texte source : anglais
LionMuon optimizer methodology - alternating between spectral (Muon) and sign-based (Lion) updates on a fixed period for compute-efficient large-scale training
Langue du texte source : anglais
Iterative algorithm to compute minimal upper bounds in the Loewner order on Hermitian matrices. Use for quantum information, convex optimization, operator theory, and numerical linear algebra tasks involving matrix inequalities.
Langue du texte source : anglais
Analyze quantum algorithms through the lens of magic (non-stabilizerness) and number-theoretic complexity. Covers the resource-theoretic framework for quantifying genuinely quantum resources in quantum algorithms, particularly Shor's factoring algorithm. Use…
Langue du texte source : anglais
Statistical learning theory methodology proving majority-of-three voting is optimal in the realizable PAC setting
Langue du texte source : anglais