All authors
Claude Skills by hiyenwong
github.com/hiyenwong9,934 skills5 installs19,223 views
- Piecewise Latent Variables For Neural Variational Text Processing**arXiv ID:** 1612.00377 **Authors:** Iulian V. Serban, Alexander G. Ororbia, Joelle Pineau, Aaron Courville **Published:** 2016-12-01T18:49:23Z **Abstract:** Advances in neural variational inference have facilitated the learning of powerful directed graphical models with continuous latent variables, such as variational autoencoders. The hope is that such models will learn to represent rich, multi-modal latent factors in real-world data, such as natural language text. However, current models ...Votes: 0GitHub stars: 3
- Planktonzilla Multimodal Dataset And Models For Understanding Plankton Ecosystems**arXiv ID:** 2606.00080 **Authors:** Alan Gerson Contreras Montanares, Luis Valenzuela, Luis Martí, Nayat Sanchez-Pi **Published:** 2026-05-22T13:54:57Z **Abstract:** Marine plankton underpin aquatic food webs and play a key role in global CO2 sequestration, making reliable species identification critical for understanding ocean health and climate feedbacks. Existing classification models perform well on individual collections but fail to generalize across instruments and environments due to...Votes: 0GitHub stars: 3
- Polydag Efficient Causal DiscoveryPolynomial acyclicity constraints for efficient continuous causal discovery in visual semantic graphs - 33% speedup over exponential baseline with improved F1 scores.Votes: 0GitHub stars: 3
- Polysemanticity And Capacity In Neural Networks**arXiv ID:** 2210.01892 **Authors:** Adam Scherlis, Kshitij Sachan, Adam S. Jermyn, Joe Benton, Buck Shlegeris **Published:** 2022-10-04T20:28:43Z **Abstract:** Individual neurons in neural networks often represent a mixture of unrelated features. This phenomenon, called polysemanticity, can make interpreting neural networks more difficult and so we aim to understand its causes. We propose doing so through the lens of feature \emph{capacity}, which is the fractional dimension each feature co...Votes: 0GitHub stars: 3
- Position As Probability Selfsupervised Transformers That Think Past Their Training For Length Extrapolation**arXiv ID:** 2506.00920 **Authors:** Philip Heejun Lee **Published:** 2025-06-01T09:20:44Z **Abstract:** Deep sequence models typically degrade in accuracy when test sequences significantly exceed their training lengths, yet many critical tasks--such as algorithmic reasoning, multi-step arithmetic, and compositional generalization--require robust length extrapolation. We introduce PRISM, a Probabilistic Relative-position Implicit Superposition Model, a novel positional encoding mechanism tha...Votes: 0GitHub stars: 3
- Pretraining With Random Noise For Uncertainty Calibration**arXiv ID:** 2412.17411 **Authors:** Jeonghwan Cheon, Se-Bum Paik **Published:** 2024-12-23T09:22:00Z **Abstract:** Uncertainty calibration is crucial for various machine learning applications, yet it remains challenging. Many models exhibit hallucinations - confident yet inaccurate responses - due to miscalibrated confidence. Here, we show that the common practice of random initialization in deep learning, often considered a standard technique, is an underlying cause of this miscalibration,...Votes: 0GitHub stars: 3
- Psvit A Methodology For Structurally Pruning Spiking Vision Transformers**arXiv ID:** 2606.03257 **Authors:** Rachmad Vidya Wicaksana Putra, Achyuta Muthuvelan, Alberto Marchisio, Muhammad Shafique **Published:** 2026-06-02T07:18:57Z **Abstract:** Spiking Vision Transformer (SViT) models are promising low-power ViT models for solving vision-based tasks with state-of-the-art performance. However, their large sizes limit their deployments for resource-constrained embedded platforms, underscoring the needs of model compression. One of prominent compression technique...Votes: 0GitHub stars: 3
- Qensemble For Offline Rl Dont Scale The Ensemble Scale The Batch Size**arXiv ID:** 2211.11092 **Authors:** Alexander Nikulin, Vladislav Kurenkov, Denis Tarasov, Dmitry Akimov, Sergey Kolesnikov **Published:** 2022-11-20T21:48:25Z **Abstract:** Training large neural networks is known to be time-consuming, with the learning duration taking days or even weeks. To address this problem, large-batch optimization was introduced. This approach demonstrated that scaling mini-batch sizes with appropriate learning rate adjustments can speed up the training process by ord...Votes: 0GitHub stars: 3
- Qif Neurons Gradient Descent AdvantageQuadratic Integrate-and-Fire (QIF) neurons exhibit continuous spike-based gradient descent with less fragmented loss landscapes and outperform LIF neurons in SNN training — computational neuroscience methodology for improved spiking neural network optimization.Votes: 0GitHub stars: 3
- Qif Superior Lif Gradient DescentQuadratic Integrate-and-Fire (QIF) neurons outperform LIF in spike-based gradient descent trainingVotes: 0GitHub stars: 3
- Qspace Novelty Detection With Variational Autoencoders**arXiv ID:** 1806.02997 **Authors:** Aleksei Vasilev, Vladimir Golkov, Marc Meissner, Ilona Lipp, Eleonora Sgarlata, Valentina Tomassini, Derek K. Jones, Daniel Cremers **Published:** 2018-06-08T07:28:36Z **Abstract:** In machine learning, novelty detection is the task of identifying novel unseen data. During training, only samples from the normal class are available. Test samples are classified as normal or abnormal by assignment of a novelty score. Here we propose novelty detection methods...Votes: 0GitHub stars: 3
- Qutrit Neural Networks Financial ForecastingQuantum qutrit-based neural network methodology for real-time financial forecasting. Uses 3-state quantum neurons instead of 2-state qubits to capture richer financial patterns with faster training.Votes: 0GitHub stars: 3
- Random Neural Network DimensionalityRandom neural networks methodology for matching observed dimensionality of neural population recordings using Dynamical Mean-Field Theory. Quantitative validation of minimal models with experimental data. Activation: 随机神经网络, 神经种群维度, dimensionality, neural population, mean-field theory, 维度性分析.Votes: 0GitHub stars: 3
- Random Projection Forest Initialization For Graph Convolutional Networks**arXiv ID:** 2302.12001 **Authors:** Mashaan Alshammari, John Stavrakakis, Adel F. Ahmed, Masahiro Takatsuka **Published:** 2023-02-22T11:49:19Z **Abstract:** Graph convolutional networks (GCNs) were a great step towards extending deep learning to unstructured data such as graphs. But GCNs still need a constructed graph to work with. To solve this problem, classical graphs such as $k$-nearest neighbor are usually used to initialize the GCN. Although it is computationally efficient to constru...Votes: 0GitHub stars: 3
- Rankingenhanced Anomaly Detection Using Active Learningassisted Attention Adversarial Dual Autoencoders**arXiv ID:** 2511.20480 **Authors:** Sidahmed Benabderrahmane, James Cheney, Talal Rahwan **Published:** 2025-11-25T16:42:12Z **Abstract:** Advanced Persistent Threats (APTs) pose a significant challenge in cybersecurity due to their stealthy and long-term nature. Modern supervised learning methods require extensive labeled data, which is often scarce in real-world cybersecurity environments. In this paper, we propose an innovative approach that leverages AutoEncoders for unsupervised anomal...Votes: 0GitHub stars: 3
- Rdnn Divisive Normalization Working MemoryRecurrent Divisive Normalization Network (RDNN) framework for continuous working memory with robust low-rank slow manifolds. Use when implementing or analyzing neural networks that need to maintain and update continuous variables without manifold shattering, particularly in computational neuroscience, working memory modeling, or RNN architecture design.Votes: 0GitHub stars: 3
- Readiness Driven Pipeline RuntimeReadiness-First Pipeline (RRFP) methodology — treating pipeline schedules as non-binding hint orders rather than pre-committed execution sequences. Reduces bubbles and stage misalignment in distributed training under runtime variability. Up to 1.77x speedup on language-only, 2.77x on multimodal workloads. Activation: readiness-driven pipeline, RRFP, pipeline parallel runtime, schedule flexibility, distributed training variability, 1F1B optimizationVotes: 0GitHub stars: 3
- Reasoning As Double Edged Sword Architecture Cross Stage Robustness Vision Language ActionSkill derived from arXiv:2607.17786 - Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-ActionVotes: 0GitHub stars: 3
- Reconfigurable Photonic Decision NetworkReconfigurable Nonlinear Photonic Decision Network (RNPDN) methodology for adaptive photonic neuromorphic computing. Local physical learning rules with tunable stability-plasticity tradeoff, controlled memory formation via bistable photonic states, and in-situ learning through driven-dissipative dynamics.Votes: 0GitHub stars: 3
- Recurrent Divisive Normalization NetworkRecurrent Divisive Normalization Network (RDNN) methodology for continuous working memory with low-rank slow manifolds. Implements biophysical divisive normalization constraint to prevent manifold shattering in RNNs while maintaining robust continuous representations. Use when: designing RNNs for continuous variable maintenance, implementing biologically-plausible working memory models, or addressing manifold shattering in artificial neural networks.Votes: 0GitHub stars: 3
- Recurrent Divisive Normalization Working MemoryRecurrent Divisive Normalization Network (RDNN) methodology for continuous working memory — uses divisive normalization to create robust low-rank slow manifolds that prevent manifold shattering under time-varying inputs. Combines biophysical constraints with gradient dynamics analysis to enable high-fidelity continuous representations in RNNs.Votes: 0GitHub stars: 3
- Recursive Gaussian Processes Predictive CodingRecursive Gaussian Processes (RGPs) methodology connecting predictive coding to Bayesian brain theories with neurobiological constraints. Use when implementing hierarchical Bayesian inference models that map to cortical microcircuits.Votes: 0GitHub stars: 3
- Recursive Kl Divergence Optimization A Dynamic Framework For Representation Learning**arXiv ID:** 2504.21707 **Authors:** Anthony D Martin **Published:** 2025-04-30T14:51:27Z **Abstract:** We propose a generalization of modern representation learning objectives by reframing them as recursive divergence alignment processes over localized conditional distributions While recent frameworks like Information Contrastive Learning I-Con unify multiple learning paradigms through KL divergence between fixed neighborhood conditionals we argue this view underplays a crucial recursive st...Votes: 0GitHub stars: 3
- Relay On Policy DistillationRelay On-Policy Distillation (Relay-OPD) methodology for trajectory-relayed token-level supervision to overcome prefix failure in reasoning models.Votes: 0GitHub stars: 3
- Renormalization Scaling Brain ActivityRenormalization group (RG) framework for analyzing scaling laws and criticality in brain activity. Connects 1/f noise, neuronal avalanches, and coarse-grained descriptions through RG theory. Activates: renormalization brain, scaling law neural activity, 1/f noise brain, neuronal avalanche scaling, coarse-graining neural dynamics, RG criticality brain, power law neural scaling.Votes: 0GitHub stars: 3
- Reopd Multi Turn On Policy DistillationReOPD (Replayed-Prefix On-Policy Distillation) methodology for scalable multi-turn agent distillation without environment interaction during training. Addresses the 'prefix trap' in multi-turn OPD via reliability-aware prefix sampling.Votes: 0GitHub stars: 3
- Representation SteeringLLM representation steering and activation patching methodology for mechanistic interpretability. Use when analyzing how steering vectors affect LLM internals, conducting activation patching experiments, or investigating causal mechanisms in neural networks. Keywords: representation steering, activation patching, mechanistic interpretability, steering vectors, OV circuit, QK circuit, refusal steering.Votes: 0GitHub stars: 3
- Representation Use Usability FrameworkUnified framework for representation use and usability across philosophy, neuroscience, cognitive science, and computer science. Analyzes when and how representations are used effectively in different systems. Trigger words: representation usability, representation use, philosophical representation, cognitive representation, AI representation theory.Votes: 0GitHub stars: 3
- Reptile A Scalable Meta Learning AlgorithmSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Reversible Sparse Moe Single NodeTrain hundred-billion-parameter sparse MoE models on single nodes using reversible recurrence stacks and state-preserving growth principles with TQP optimizer strategy.Votes: 0GitHub stars: 3
- Riemannian Retrieval Fisher DistillationInformation-geometry unified memory architecture combining Riemannian retrieval (Fisher-Rao metric) with Fisher-guided discrete token distillation for resource-efficient long-term memory in dialogue agents.Votes: 0GitHub stars: 3
- Risk Bounds For Robust Deep Learning**arXiv ID:** 2009.06202 **Authors:** Johannes Lederer **Published:** 2020-09-14T05:06:59Z **Abstract:** It has been observed that certain loss functions can render deep-learning pipelines robust against flaws in the data. In this paper, we support these empirical findings with statistical theory. We especially show that empirical-risk minimization with unbounded, Lipschitz-continuous loss functions, such as the least-absolute deviation loss, Huber loss, Cauchy loss, and Tukey's biweight loss...Votes: 0GitHub stars: 3
- Rlcsd Contrastive On Policy DistillationContrastive on-policy self-distillation methodology for reasoning models that mitigates privilege-induced style driftVotes: 0GitHub stars: 3
- Rlvp Penalize The Path Reward The OutcomeAgents acting on our behalf in the real world (e.g. placing phone calls) must learn online from costly, often irreversible interactions rather than cheap simulator steps. Two things follow. First, dep. Based on arXiv:2607.07435.Votes: 0GitHub stars: 3
- Rolling The Dice For Better Deep Learning Performance A Study Of Randomness Techniques In Deep Neural Networks**arXiv ID:** 2404.03992 **Authors:** Mohammed Ghaith Altarabichi, Sławomir Nowaczyk, Sepideh Pashami, Peyman Sheikholharam Mashhadi, Julia Handl **Published:** 2024-04-05T10:02:32Z **Abstract:** This paper investigates how various randomization techniques impact Deep Neural Networks (DNNs). Randomization, like weight noise and dropout, aids in reducing overfitting and enhancing generalization, but their interactions are poorly understood. The study categorizes randomness techniques into four...Votes: 0GitHub stars: 3
- Rollout Adaptive Supervised Finetuning RasftRollout-Adaptive Supervised Fine-Tuning (RASFT) for reasoning tasks - policy-aware SFT that calibrates expert supervision based on problem-level solvability from verified rollouts.Votes: 0GitHub stars: 3
- Rp Opsd Reasoning Pivot DistillationRP-OPSD for reasoning-pivot-guided distillation.Votes: 0GitHub stars: 3
- Saber Spatial Attention Brain XrSABER framework integrating spatial attention neuroscience with Extended Reality for adaptive human-computer interaction. Activation: spatial attention XR, brain-computer interface, attention-aware computing, extended reality neuroscience, eye-tracking optimization.Votes: 0GitHub stars: 3
- Sae Optimality Structures DictionariesSAE 最优性结构理论 - 解释 Sparse Autoencoders 如何从最优性条件提取可解释特征。涵盖层次分裂与吸收、残差结构、密集对立特征等现象的理论基础。Votes: 0GitHub stars: 3
- Sae Optimality StructuresTheory explaining how optimality conditions structure SAE (Sparse Autoencoder) dictionaries - hierarchical splitting, absorption, residuals, and dense antipodal featuresVotes: 0GitHub stars: 3
- Sampling On Random Subspaces Under Limited Data InSampling on Random Subspaces under Limited Data in the Context of Exploratory Landscape Analysis. Classical space-filling designs often fail to provide reliable statistical results for Exploratory Landscape Analysis (ELA) when only limited evaluation budgets are available, as commonly occurs in hi... Activation: benchmark, optimization, lora, robustness, embeddingVotes: 0GitHub stars: 3
- Scalable Memristive Reservoir ComputingScalable Memristive-Friendly Reservoir Computing for time series classification using Memristive-Friendly Echo State Networks (MF-ESN). Combines memristive device physics with reservoir computing for efficient time-series classification. Trigger words: memristive reservoir computing, MF-ESN, echo state network, memristor time series, hardware reservoir, memristive ESN, time series classification, reservoir computing classification.Votes: 0GitHub stars: 3
- Scalable Qst Neural ArchitecturesBenchmarking neural network architectures for scalable Quantum State Tomography with memristor-based acceleration patternsVotes: 0GitHub stars: 3
- Selective Timestep Weighting And Advantage Based Replay For Sample EfficientReinforcement learning from human feedback (RLHF) has emerged as a powerful paradigm for aligning generative models with human preferences. However, applying RLHF to diffusion models remains highly fe. Based on arXiv:2607.07693.Votes: 0GitHub stars: 3
- Self ChallengeSelf-evolution skill that uses dual-agent challenge design and execution to expand capabilities over time.Votes: 0GitHub stars: 3
- Self Initiated Attention Shifts EegSubject-specific analysis of self-initiated attention shifts from EEG with controlled internal and external attention conditions. Machine learning + SHAP feature attribution reveals that higher-frequency bands and frontal regions carry subject-specific discriminative information for distinguishing self-initiated vs externally-cued attention shifts (arXiv:2605.18251). Use for EEG attention decoding, self-initiated attention research, voluntary attention neural correlates, SHAP-based EEG interp...Votes: 0GitHub stars: 3
- Self Verification自我验证技能，基于 ReVeal 论文实现多轮生成-验证迭代，支持代码和推理任务的可靠自我验证。触发词：自我验证、self-verification、verify、验证代码、验证推理。Votes: 0GitHub stars: 3
- Semidefinite Programming Causal GamesGPU-accelerated semidefinite programming for causal game analysis — using SDP hierarchies to compute bounds in causal inference games, with GPU acceleration for scalability. From arXiv:2606.20519.Votes: 0GitHub stars: 3
- Sevalnas A Searchagnostic Evaluation For Neural Architecture Search**arXiv ID:** 2603.00099 **Authors:** Atah Nuh Mih, Jianzhou Wang, Truong Thanh Hung Nguyen, Hung Cao **Published:** 2026-02-17T15:02:02Z **Abstract:** Neural architecture search (NAS) automates the discovery of neural networks that meet specified criteria, yet its evaluation procedures are often hardcoded, limiting the ability to introduce new metrics. This issue is especially pronounced in hardware-aware NAS, where objectives depend on target devices such as edge hardware. To address this l...Votes: 0GitHub stars: 3
- Shia Sysml Hardware InterfaceSysML-Hardware Interface Architecture (SHIA) for model-centric verification. Keeps executable SysML models directly in the hardware verification loop without intermediate transformations. Use when: designing model-to-hardware interfaces, implementing hardware-in-the-loop (HiL) verification with MBSE/SysML, building digital threads between system models and physical hardware, or integrating SysML models with embedded systems for V&V. Keywords: SHIA, SysML, MBSE, hardware-in-the-loop, HiL, mode...Votes: 0GitHub stars: 3