All authors
Claude Skills by hiyenwong
github.com/hiyenwong9,934 skills5 installs19,223 views
- Tta Eeg Foundation ModelsTest-time adaptation (TTA) methods for EEG foundation models under real-world distribution shifts. Covers entropy minimization, batch norm adaptation, TENT, MEMO, CoTTA, EATA, SAR, and related methods applied to EEG signals across cross-device, cross-subject, cross-session, and cross-clinical-site settings. Addresses privacy-preserving adaptation without source data access.Votes: 0GitHub stars: 3
- Ultrasound Neuromodulation Prediction FrameworkOpen-source computational framework for predicting ultrasound neuromodulation effects by bridging tissue elastomechanics and neuron firing dynamics. Maps transcranial acoustic fields to per-voxel neural firing maps using coupled physics models. Activation: ultrasound neuromodulation, transcranial focused ultrasound, neural firing prediction, tissue elastomechanics, Hodgkin-Huxley neuronVotes: 0GitHub stars: 3
- Unibci Invasive Foundation ModelUniBCI: unified pretrained foundation model for invasive Brain-Computer Interfaces. Context-conditioned spatio-temporal tokenization, Interval-Area Attention, self-supervised masked reconstruction. Trigger words: UniBCI, invasive BCI, neural spike foundation model, brain-computer interface, spatio-temporal tokenization.Votes: 0GitHub stars: 3
- Unifying Von Neumann Hpc Neuromorphic Ebbrains统一冯诺依曼HPC与神经形态计算的EBRAINS工作流框架 - 透明跨平台执行SNN，trigger_words: neuromorphic computing. spiking neural network. SNN. HPC. EBRAINS. SpiNNaker. cross-platform. containerization. NESTMLVotes: 0GitHub stars: 3
- Unispike Snn AccelerationUniSpike methodology for accelerating Spiking Neural Networks (SNNs) on many-core neuromorphic systems by eliminating address redundancy in packet-based spike communication. Combines destination-centric spike scheduling, lightweight runtime packet assembly hardware, and destination-aware SNN partitioning to reduce traffic, improve speed, and increase energy efficiency. Use when: (1) optimizing SNN inference on neuromorphic hardware, (2) designing spike communication protocols for many-core ne...Votes: 0GitHub stars: 3
- Universal Bci Personalization ApiUniversal BCI Personalization API for trunk-agnostic EEG foundation model integration. Provides one contract encode to Bayesian head to BrainState architecture that works across heterogeneous frozen EEG trunks without per-architecture personalization stacks. Use when implementing BCI systems that need to support multiple EEG encoder architectures (EEGNet, Shallow, Deep, Conformer, ATCNet, REVE) with a single personalization interface.Votes: 0GitHub stars: 3
- Universal Brain DynamicsUniversal Brain Dynamics (UBD) methodology for constructing a universal latent space of brain activity that integrates structural connectivity (dMRI) with temporal dynamics (fMRI) using GCNs and Deep Koopman Operators. Achieves Pearson's r > 0.9 across 8 cognitive states and 963 subjects. Enables analysis of cognitive transitions, structure-function coupling, and individual differences. Use when: (1) analyzing whole-brain fMRI dynamics, (2) studying structure-function coupling in the brain, (...Votes: 0GitHub stars: 3
- Universal Neural Propagator Quantum DynamicsUniversal Neural Propagator (UNP) methodology for learning time evolution in many-body quantum systems. Transfers across both Hamiltonians and initial states simultaneously. Activation: neural propagator, quantum dynamics simulation, neural operator learning, quantum state evolution, UNP, universal propagator, quantum foundation model, neural quantum dynamics.Votes: 0GitHub stars: 3
- Universal Neural PropagatorUniversal Neural Propagator (UNP) methodology for learning time evolution in many-body quantum systems. A single self-supervised model that maps driving protocols to time-evolution propagators, predicting dynamics across a function space of driving protocols and an exponentially large Hilbert space of initial states simultaneously. Use when: (1) modeling quantum system time evolution under varying Hamiltonians, (2) learning propagator mappings from driving protocols, (3) self-supervised train...Votes: 0GitHub stars: 3
- Universal Object Representations VisionDecomposition methodology for identifying universal vs model-specific dimensions in vision model representations across 162 diverse models. Universal dimensions are more interpretable, driven by conceptual image properties, and better predict macaque IT activity and human similarity judgments. arXiv:2605.13675.Votes: 0GitHub stars: 3
- Unsupervised Behaviour Discovery With Qualitydiversity Optimisation**arXiv ID:** 2106.05648 **Authors:** Luca Grillotti, Antoine Cully **Published:** 2021-06-10T10:40:18Z **Abstract:** Quality-Diversity algorithms refer to a class of evolutionary algorithms designed to find a collection of diverse and high-performing solutions to a given problem. In robotics, such algorithms can be used for generating a collection of controllers covering most of the possible behaviours of a robot. To do so, these algorithms associate a behavioural descriptor to each of these...Votes: 0GitHub stars: 3
- Unsupervised Oneshot Learning Of Both Specific Instances And Generalised Classes With A Hippocampal Architecture**arXiv ID:** 2010.15999 **Authors:** Gideon Kowadlo, Abdelrahman Ahmed, David Rawlinson **Published:** 2020-10-30T00:10:23Z **Abstract:** Established experimental procedures for one-shot machine learning do not test the ability to learn or remember specific instances of classes, a key feature of animal intelligence. Distinguishing specific instances is necessary for many real-world tasks, such as remembering which cup belongs to you. Generalisation within classes conflicts with the ability t...Votes: 0GitHub stars: 3
- Updated Neuron Model AnnUpdating the standard neuron model in artificial neural networks — replacing point neuron model with realistic cortical cell model improves expressivity, robustness, and learning efficiencyVotes: 0GitHub stars: 3
- User As Engram Hippocampal Memory Architecture大脑启发式记忆架构方法论 - 将用户记忆分离为内容层（海马体式 engram）和技能层（新皮层式共享 adapter），实现高效个人化 LLMVotes: 0GitHub stars: 3
- Using Hierarchical Statistical Learning Models To Model IndividualUsing hierarchical statistical learning models to model individual statistical learning. Statistical learning is essential for individuals to discover structure in the sensory environment, especially during communication via speech or music. Individual differences in statistical learning ... Activation: graph, cognitive, eeg, communication, bayesianVotes: 0GitHub stars: 3
- V1 Digital Twin Latent ProbingMulti-level representational probing framework for digital twins of mouse V1. Systematically probes latent representations in neural activity-predicting models across three levels: (i) linear decodability from controlled visual probes, (ii) latent-unit tuning to canonical visual features, (iii) population geometry of hidden-layer activity. Reveals that digital twins with comparable prediction accuracy can differ substantially in internal representations. Activation: digital twin probing, V1 l...Votes: 0GitHub stars: 3
- V1 Digital Twin ProbingBeyond Neural Activity Prediction: Probing Latent Representations in Mouse V1 Digital Twins. Systematic multi-level probing framework for evaluating latent representations in sensory cortex digital twins, across linear decodability, latent-unit tuning, and population geometry. Activation: V1 digital twin, latent representation probing, neural prediction, population geometry, representational similarity, mouse V1 encoding, digital twin evaluation, sensory cortex modelingVotes: 0GitHub stars: 3
- Vacoal Hippocampal MemoryVaCoAl algebro-deterministic hippocampal memory architecture. Bridges silicon and hippocampus by connecting Vector-HaSH, TEM, and hyperdimensional computing. Use when: hippocampal memory modeling, grid-cell scaffolds, episodic memory replay, sharp-wave ripples, Vector-HaSH, Tolman-Eichenbaum Machine, hyperdimensional computing, Galois-field LFSRs, multi-hop memory replay, memory fidelity decay, STDP-like path selection, EC-CA3 pathways, dentate gyrus, compositional memory, hippocampal-entorhi...Votes: 0GitHub stars: 3
- Vae Quantum EmbeddingVariational Autoencoder (VAE) framework for learning task-specific quantum embeddings of classical data. Compresses high-dimensional datasets (including ImageNet) into compact quantum representations (e.g., 13-qubit) while remaining reconstructable through a learned decoder. Achieves polynomial-measurement reconstruction (vs. full tomography for amplitude embeddings or circuit inversion for angle embeddings). Validated on IBM quantum hardware with stable embeddings under real device noise. Us...Votes: 0GitHub stars: 3
- Valence Axis Llm Eeg Saturation RegularityA Shared Valence Axis Across Modern LLMs and Human EEG: The Saturation Regularity (arXiv:2606.00129). LLM-derived valence direction maps onto human EEG, revealing saturation regularity: task supervision saturates basin, additional alignment distorts residual. Ensemble across residual diversity improves decoding by 10.5%. Activation: valence axis, LLM EEG alignment, saturation regularity, emotional valence decoding, brain-language model alignment, residual ensemble, EEG emotion classification.Votes: 0GitHub stars: 3
- Valence Bond Embedding Quantum ChemistryValence bond embedding methodology for mapping deep quantum chemistry computations onto shallow quantum circuits, reducing NISQ resource requirements.Votes: 0GitHub stars: 3
- Variance Brain Foundation Models Forgot脑基础模型方差分配问题方法论。揭示BFMs预测认知失败的根本原因 - 预训练捕获主要方差成分但丢失三阶统计量（协偏度）。线性协偏度子空间FC方法超越所有BFMs（无预训练、无GPU）。规模悖论：BrainLM 650M预测认知比111M更差。适用于脑基础模型评估、认知预测、fMRI分析、统计量保留。触发词：brain foundation models、BFM、variance allocation、third-order statistics、co-skewness、cognition prediction、脑基础模型方差、协偏度、认知预测失败。Votes: 0GitHub stars: 3
- Variational Autoencoder Learns Better Feature Representations For Eegbased Obesity Classification**arXiv ID:** 2302.00789 **Authors:** Yuan Yue, Jeremiah D. Deng, Dirk De Ridder, Patrick Manning, Divya Adhia **Published:** 2023-02-01T22:48:45Z **Abstract:** Obesity is a common issue in modern societies today that can lead to various diseases and significantly reduced quality of life. Currently, research has been conducted to investigate resting state EEG (electroencephalogram) signals with an aim to identify possible neurological characteristics associated with obesity. In this study, we...Votes: 0GitHub stars: 3
- Variational Phasor Circuits BciVariational Phasor Circuits (VPC) for phase-native Brain-Computer Interface classification using continuous S1 unit circle manifold with trainable phase shifts and unitary mixingVotes: 0GitHub stars: 3
- Vau Da Muntanialas Energyefficient Multidie Scalable Acceleration Of Rnn Inference**arXiv ID:** 2202.07462 **Authors:** Gianna Paulin, Francesco Conti, Lukas Cavigelli, Luca Benini **Published:** 2022-02-14T09:21:16Z **Abstract:** Recurrent neural networks such as Long Short-Term Memories (LSTMs) learn temporal dependencies by keeping an internal state, making them ideal for time-series problems such as speech recognition. However, the output-to-input feedback creates distinctive memory bandwidth and scalability challenges in designing accelerators for RNNs. We present Mun...Votes: 0GitHub stars: 3
- Veclstm Trajectory Data Processing And Management For Activity Recognition Through Lstm Vectorization And Database Integration**arXiv ID:** 2409.19258 **Authors:** Solmaz Seyed Monir, Dongfang Zhao **Published:** 2024-09-28T06:22:44Z **Abstract:** Activity recognition is a challenging task due to the large scale of trajectory data and the need for prompt and efficient processing. Existing methods have attempted to mitigate this problem by employing traditional LSTM architectures, but these approaches often suffer from inefficiencies in processing large datasets. In response to this challenge, we propose VecLSTM, a n...Votes: 0GitHub stars: 3
- Vector Space Of Cycles Harmonic FlowVariational framework for statistical inference on cyclic interactions in directed networks. Directed interactions as edge flows on simplicial complex evolved under energy-minimizing dynamics, yielding low-dimensional cycle space for recurrent organization. Activation: cyclic interaction, harmonic flow, cycle space, simplicial complex, recurrent network, directed graph cycles.Votes: 0GitHub stars: 3
- Ven Circuit Snn Social LearningVENCircuit methodology — Von Economo neurons (VENs) as acquisition scaffolds in recurrent spiking neural networks. Combines computational modeling with clinical predictions for bvFTD and autism. Use when: studying VENs, social learning in SNNs, gradient pathway analysis, clinical prediction from computational models, developmental scaffolding in neural networks.Votes: 0GitHub stars: 3
- Vencircuit Ven Scaffold SnnComputational framework showing Von Economo neurons (VENs) function as acquisition scaffolds in recurrent spiking neural networks. Provides gradient pathways immune to Jacobian instabilities, explaining VEN loss in bvFTD and ASC.Votes: 0GitHub stars: 3
- Vencircuit Von Economo Snn Social LearningVENCircuit methodology — computational account of Von Economo neurons (VENs) as acquisition scaffolds in recurrent spiking neural networks for reliable social skill learning. Use when analyzing VEN function in SNN training, gradient-flow theory for residual pathways in recurrent circuits, or computational models of autism spectrum conditions (ASC) and frontotemporal dementia (bvFTD).Votes: 0GitHub stars: 3
- Verievol Verifiable Data ConstructionVerifiable Evol-Instruct framework for scaling multimodal mathematical reasoning. Type-aware evolution + HTV-Agent verifier with offline hypothesis-test falsification ensures reliable reward labels at scale.Votes: 0GitHub stars: 3
- Visae Neuroscience Concept Circuits VitViSAE (Visual Sparse Autoencoder for Interpretability) - Neuroscience-motivated concept circuits framework for interpreting and steering Vision Transformers. Uses 64K images with 16K visually grounded concept vocabulary, achieving 20x efficiency improvement and 28.7% accuracy boost. Provides top-down concept reading and bottom-up circuit tracing algorithms for automated interpretability. ICML 2026 paper.Votes: 0GitHub stars: 3
- Vision Hopfield Memory NetworksVision Hopfield Memory Network (V-HMN) - brain-inspired backbone with hierarchical Hopfield memory + predictive-coding refinement. Local patch memory + global episodic memory + error correction. Enhanced interpretability, data efficiency, biological plausibility. Memory modules replace self-attention/state-space, exposing input-pattern relationships. Use for: brain-inspired vision, interpretable backbones, data-efficient training, associative memory, Transformer/Mamba alternatives. Activation...Votes: 0GitHub stars: 3
- Visual Cortex Diffusion InferenceSkill for understanding and applying the mechanistic model of perceptual inference in visual cortex equivalent to a minimal diffusion model (arXiv:2607.15693). Enables extraction of principles linking sparse coding, recurrent dynamics, and diffusion model training for neuroscience-inspired machine learning.Votes: 0GitHub stars: 3
- Visual Cortex Diffusion Model--- name: visual-cortex-diffusion-model description: "Skill for understanding and applying the mechanistic model of inference in visual cortex equivalent to a minimal diffusion model, linking sparse coding with recurrent dynamics and horizontal connections in V1. Based on arXiv:2607.15693." activation: visual cortex diffusion model, sparse coding inference, recurrent diffusion modelVotes: 0GitHub stars: 3
- Visual Imagery Decoding FmriLatent functional alignment approach for decoding visual imagery from fMRI data. Extends perception-optimized pipelines to mental imagery using diffusion-based generative models. Activation: fmri, decoding, visual-imagery, diffusion-models, functional-alignment, neuroscience, brain, neuralVotes: 0GitHub stars: 3
- Visual Place Recognition Rate Encoded Snn StdpSkill for understanding and implementing the discrete tensor-native STDP-based SNN visual place recognition pipeline from arXiv:2607.13584v1. Use when working with spiking neural networks for visual place recognition, loop closure in SLAM, or neuromorphic computing applications.Votes: 0GitHub stars: 3
- Visual Semantic Decoding Ecog VideoEnd-to-end deep learning framework for visual semantic decoding from ECoG, demonstrating promising performance without handcrafted features while maintaining interpretability.Votes: 0GitHub stars: 3
- Vlm Lam Brain AlignmentBrain alignment methodology for comparing Vision-Language Models (VLMs) and Large-Action Models (LAMs) against human fMRI during naturalistic interactive tasks. Studies how reasoning-focused vs action-focused prompts shape model internal representations and their alignment with brain activity across the cortical hierarchy. Use when: evaluating interactive AI model brain alignment, comparing VLM vs LAM neural representations, studying prompt-driven representation changes, designing fMRI encodi...Votes: 0GitHub stars: 3
- Vlm Visual Cortex Alignment RobustnessVisual Language Model robustness through early visual cortex alignment. Reveals V1-V3 alignment improves VLM resistance to adversarial manipulation. Triggers: VLM robustness, visual cortex alignment, adversarial defense, neuroscience AI, V1-V3 alignment, sycophancy prevention.Votes: 0GitHub stars: 3
- Vo2 Mott Oscillator Spiking Neurons V2Monolithically integrated VO2 Mott phase-transition oscillators for compact, energy-efficient hardware implementation of spiking neurons. Compatible with large-scale CMOS integration for brain-inspired non-Boolean computing.Votes: 0GitHub stars: 3
- Vo2 Mott Oscillator Spiking NeuronsMonolithically integrated VO2 Mott oscillators for energy-efficient spiking neurons. Metal-insulator transition devices for neuromorphic hardware. Activation: VO2 oscillator, Mott neuron, metal-insulator transition, energy-efficient spiking.Votes: 0GitHub stars: 3
- Voice Memory For Agentic Speech RecognitionVoice Memory for Agentic Speech RecognitionVotes: 0GitHub stars: 3
- Von Economo Fast Lane HypothesisVon 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.Votes: 0GitHub stars: 3
- Vs Wno Variable Spiking WaveletVariable Spiking Wavelet Neural Operator (VS-WNO) — a systematic study of spiking sparsity versus real-world deployment cost on edge GPUs. Covers wavelet neural operators augmented with spiking mechanisms, variable sparsity control, hardware-aware model design, and deployment profiling on NVIDIA Jetson Orin Nano 8GB.Votes: 0GitHub stars: 3
- Warmstart Dl Unit CommitmentMulti-Stage Warm-Start (MSWS) deep learning framework for Unit Commitment optimization. Combines neural network warm-starting with MILP constraints to accelerate power grid scheduling. Use for unit commitment, power system optimization, energy scheduling, and MILP warm-starting.Votes: 0GitHub stars: 3
- Wavelet Scattering Schizophrenia Eeg BiomarkerWavelet Scattering Transform (WST) framework for interpretable schizophrenia biomarker discovery and classification from resting-state EEG. Multi-order scattering coefficients capture cross-frequency coupling and amplitude modulation dynamics, achieving 90.48% accuracy under strict subject-independent evaluation.Votes: 0GitHub stars: 3
- Weight Geometry Functional MemoryUniversal organizational regularity showing weight geometry governs functional memory in complex systems across biological, ecological, social, and technological domains. Activation: weight geometry, functional memory, complex systems, interaction strength, hierarchical organization.Votes: 0GitHub stars: 3
- Weighted Brain Community Detection加权脑网络社区检测方法论，突破分辨率限制。 使用 Asymptotical Surprise 检测多尺度功能模块。 触发词：脑网络、社区检测、模块化、分辨率限制、Surprise、 community detection, weighted network, brain connectivity, modular organization。Votes: 0GitHub stars: 3
- Weighted Partitions Interval RestrictionsExact formulas and bivariate master identity methodology for weighted partitions with interval restrictions. Use when: deriving generating functions for restricted partition functions, proving partition coefficient bounds via Rogers-Fine evaluation, establishing master identities with auxiliary variables, or analyzing quantum modular forms from interval-restricted partitions. Activation: weighted partitions, interval restrictions, bivariate master identity, Rogers-Fine evaluation, false theta...Votes: 0GitHub stars: 3