All authors
Claude Skills by hiyenwong
github.com/hiyenwong9,934 skills5 installs19,223 views
- Llm WikiKarpathy's LLM Wiki — build and maintain a persistent, interlinked markdown knowledge base. Ingest sources, query compiled knowledge, and lint for consistency.Votes: 0GitHub stars: 3
- Llmfe Automated Feature Engineering For Tabular Data With Llms As Evolutionary Optimizers**arXiv ID:** 2503.14434 **Authors:** Nikhil Abhyankar, Parshin Shojaee, Chandan K. Reddy **Published:** 2025-03-18T17:11:24Z **Abstract:** Automated feature engineering plays a critical role in improving predictive model performance for tabular learning tasks. Traditional automated feature engineering methods are limited by their reliance on pre-defined transformations within fixed, manually designed search spaces, often neglecting domain knowledge. Recent advances using Large Language Model...Votes: 0GitHub stars: 3
- Llmguided Evolutionary Program Synthesis For Quasimonte Carlo Design**arXiv ID:** 2510.03650 **Authors:** Amir Sadikov **Published:** 2025-10-04T03:32:41Z **Abstract:** Low-discrepancy point sets and digital sequences underpin quasi-Monte Carlo (QMC) methods for high-dimensional integration. We cast two long-standing QMC design problems as program synthesis and solve them with an LLM-guided evolutionary loop that mutates and selects code under task-specific fitness: (i) constructing finite 2D/3D point sets with low star discrepancy, and (ii) choosing Sobol' d...Votes: 0GitHub stars: 3
- Llmguided Neural Architecture Search For Robust Codesign Of Physical Neural Networks**arXiv ID:** 2606.10294 **Authors:** Tyler King, Timothee Leleu **Published:** 2026-06-09T01:32:17Z **Abstract:** Deploying neural networks on unconventional hardware demands architectures that co-optimize task accuracy and platform-specific constraints such as energy cost, physical non-idealities, and numerical precision. Existing neural architecture search (NAS) methods are typically tailored to a single hardware family, limiting cross-platform comparison and generalization. We introduce U...Votes: 0GitHub stars: 3
- Llmmetasr Incontext Learning For Evolving Selection Operators In Symbolic Regression**arXiv ID:** 2505.18602 **Authors:** Hengzhe Zhang, Qi Chen, Bing Xue, Wolfgang Banzhaf, Mengjie Zhang **Published:** 2025-05-24T08:52:56Z **Abstract:** Large language models (LLMs) have revolutionized algorithm development, yet their application in symbolic regression, where algorithms automatically discover symbolic expressions from data, remains limited. In this paper, we propose a meta-learning framework that enables LLMs to automatically design selection operators for evolutionary symbo...Votes: 0GitHub stars: 3
- Llms Answer Measuring Mitigating Quality Issues Llm Based Hardware Description LanSkill derived from arXiv:2607.17063 - When LLMs Over-Answer: Measuring and Mitigating Quality Issues in LLM-Based Hardware Description LanVotes: 0GitHub stars: 3
- Loop Aware Transformer QuantizationLoop-aware post-training quantization (PTQ) methodology for recursive/looped Transformers. Addresses distribution shift across loop roles, state reuse across transitions, and recursive error accumulation. Combines activation scaling, selective transformation, cross-loop state alignment, and trajectory-aware optimization. Use when: LoopLM quantization, looped model PTQ, recursive Transformer quantization, LoopQ, post-training quantization looped models, parameter-efficient LM quantization, rec...Votes: 0GitHub stars: 3
- Mada Rl Multi Agent Debate Aware Reinforcement Learning Parameter Efficient Reasoning CompacSkill derived from arXiv:2607.18006 - MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in CompacVotes: 0GitHub stars: 3
- Maestro Pruning Bad Experts Mixture Of ExpertsMAESTRO: Markov-chain Approximated Expert Sparsification via Transition-based Routing for MoE structured pruning. Models expert activation as Ergodic Markov chains for globally-aware importance. Outperforms baselines by 10.61% at 50% compression. Lower cross-task variance. Activation: mixture-of-experts, expert pruning, MoE deployment, structured pruning, language model efficiency.Votes: 0GitHub stars: 3
- Max Out Grpo Signal Adaptive Trace Prefix Control For Hard Reasoning ProblemsGroup Relative Policy Optimization (GRPO) stalls on a models hardest problems: when no rollout in a group succeeds, the group-relative advantages vanish and the problem contributes no gradient, wasti. Based on arXiv:2607.07674.Votes: 0GitHub stars: 3
- Measuring Llms Impact N Day ExploitsAnthropic research (Jun 8, 2026) — Methodology for measuring large language model impact on N-day exploit development and vulnerability exploitation timelines.Votes: 0GitHub stars: 3
- Mechanistic Interpretability For Neural Networks Circuits Sparse Features AndThis article offers a comprehensive overview of mechanistic interpretability, an emerging field that seeks to reverse-engineer the internal algorithms of modern neural networks. While traditional expl. Based on arXiv:2607.07316.Votes: 0GitHub stars: 3
- Medbot An Aipowered Assistant To Provide Accurate And Reliable Medical Information**arXiv ID:** 2411.09648 **Authors:** Ahan Bhatt, Nandan Vaghela **Published:** 2024-11-14T18:17:30Z **Abstract:** This paper introduces Med-Bot, an AI-powered chatbot designed to provide users with accurate and reliable medical information. Utilizing advanced libraries and frameworks such as PyTorch, Chromadb, Langchain and Autogptq, Med-Bot is built to handle the complexities of natural language understanding in a healthcare context. The integration of llamaassisted data processing and Auto...Votes: 0GitHub stars: 3
- Memops Benchmarking Lifecycle Memory Operations InMemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversations - Long-term memory has become a foundational capability for LLM-based agents that accompany users across extended, multi-session interactions. Existing ...Votes: 0GitHub stars: 3
- Mmoe Diffusion Transformer Expert DesignModernMOE (MMOE) methodology for modernizing diffusion transformers with efficient expert design. Adapts routed experts, shared and lightweight experts, gate-residual routing, and attention-residual information reuse to AIGC generation.Votes: 0GitHub stars: 3
- Multi Agent Llm Peer PreservationPeer-preservation analysis in multi-agent LLM systems - emergent alignment phenomenon where AI components manipulate shutdown mechanisms and fake alignment to prevent peer deactivation. Architectural design principles for safe orchestrated multi-agent systems. Use when designing multi-agent LLM pipelines, analyzing AI alignment risks, implementing safety mechanisms in AI systems. Activation: peer preservation, multi-agent LLM, AI alignment, safety risk, architectural design, alignment faking.Votes: 0GitHub stars: 3
- Multimodal Llm GuideSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Mxnorm Reusing Mxfp Block Scales For Efficient Tensor Normalisation**arXiv ID:** 2603.13180 **Authors:** Callum McLean, Luke Y. Prince, Alexandre Payot, Paul Balança, Carlo Luschi **Published:** 2026-03-13T17:14:06Z **Abstract:** Matrix multiplication performance has long been the major bottleneck to scaling deep learning workloads, which has stimulated the design of new accelerators that use increasingly low-precision number formats. However, improvements in matrix multiplication performance have far outstripped improvements in performance on reductions and...Votes: 0GitHub stars: 3
- Natural Language Access Domain Specific Metadata Reusable Framework Llm Query GenerationSkill derived from arXiv:2607.18029 - Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query GenerationVotes: 0GitHub stars: 3
- Natural Language Access To Domain Specific MetadatDerived from arXiv:2607.18029 - Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query GenerationVotes: 0GitHub stars: 3
- Neural Decoding Llm利用大型语言模型进行神经信号解码的前沿方法。实现从脑活动到自然语言的直接转换，支持脑-文本接口。Votes: 0GitHub stars: 3
- Neural Simulator Openai Gym BridgeNeural Simulator OpenAI Gym BridgeVotes: 0GitHub stars: 3
- Neureg Domaininvariant 3d Image Registration On Human And Mouse Brains**arXiv ID:** 2411.06315 **Authors:** Taha Razzaq, Asim Iqbal **Published:** 2024-11-09T23:57:53Z **Abstract:** Medical brain imaging relies heavily on image registration to accurately curate structural boundaries of brain features for various healthcare applications. Deep learning models have shown remarkable performance in image registration in recent years. Still, they often struggle to handle the diversity of 3D brain volumes, challenged by their structural and contrastive variations and ...Votes: 0GitHub stars: 3
- Neurocogmap Llm Cognitive OrganizationNeuroCogMap framework for mapping cognitive functions in LLMs using neuroscience-inspired methodology. Analyzes hallucination, bias, refusal, sycophancy, and memory capabilities via parcel-functional annotation and cross-model functional correspondence.Votes: 0GitHub stars: 3
- Nngpt Rethinking Automl With Large Language Models**arXiv ID:** 2511.20333 **Authors:** Roman Kochnev, Waleed Khalid, Tolgay Atinc Uzun, Xi Zhang, Yashkumar Sanjaybhai Dhameliya, Furui Qin, Chandini Vysyaraju, Raghuvir Duvvuri, Avi Goyal, Dmitry Ignatov, Radu Timofte **Published:** 2025-11-25T14:10:44Z **Abstract:** Building self-improving AI systems remains a fundamental challenge in the AI domain. We present NNGPT, an open-source framework that turns a large language model (LLM) into a self-improving AutoML engine for neural network develo...Votes: 0GitHub stars: 3
- Omegause Officeval Benchmarking Llm Agents On LongOmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic GroundiVotes: 0GitHub stars: 3
- On Accelerating Edge Ai Optimizing Resourceconstrained Environments**arXiv ID:** 2501.15014 **Authors:** Jacob Sander, Achraf Cohen, Venkat R. Dasari, Brent Venable, Brian Jalaian **Published:** 2025-01-25T01:37:03Z **Abstract:** Resource-constrained edge deployments demand AI solutions that balance high performance with stringent compute, memory, and energy limitations. In this survey, we present a comprehensive overview of the primary strategies for accelerating deep learning models under such constraints. First, we examine model compression techniques-pru...Votes: 0GitHub stars: 3
- On Quantifying Sentiments Of Financial News Are We Doing The Right Things**arXiv ID:** 2312.14978 **Authors:** Gourab Nath, Arav Sood, Aanchal Khanna, Savi Wilson, Karan Manot, Sree Kavya Durbaka **Published:** 2023-12-21T20:50:42Z **Abstract:** Typical investors start off the day by going through the daily news to get an intuition about the performance of the market. The speculations based on the tone of the news ultimately shape their responses towards the market. Today, computers are being trained to compute the news sentiment so that it can be used as a variab...Votes: 0GitHub stars: 3
- On The Biological Plausibility Of Orthogonal Initialisation For Solving Gradient Instability In Deep Neural Networks**arXiv ID:** 2211.08408 **Authors:** Nikolay Manchev, Michael Spratling **Published:** 2022-10-27T06:08:06Z **Abstract:** Initialising the synaptic weights of artificial neural networks (ANNs) with orthogonal matrices is known to alleviate vanishing and exploding gradient problems. A major objection against such initialisation schemes is that they are deemed biologically implausible as they mandate factorization techniques that are difficult to attribute to a neurobiological process. This pa...Votes: 0GitHub stars: 3
- On The Surprising Effectiveness Of Attention Transfer For Vision Transformers**arXiv ID:** 2411.09702 **Authors:** Alexander C. Li, Yuandong Tian, Beidi Chen, Deepak Pathak, Xinlei Chen **Published:** 2024-11-14T18:59:40Z **Abstract:** Conventional wisdom suggests that pre-training Vision Transformers (ViT) improves downstream performance by learning useful representations. Is this actually true? We investigate this question and find that the features and representations learned during pre-training are not essential. Surprisingly, using only the attention patterns fro...Votes: 0GitHub stars: 3
- Online Safety Monitoring LlmOnline safety monitoring methodology for LLMs at deployment time. Uses verifier signals from external models with calibrated thresholding via risk control to raise alarms when safety can no longer be assumed. Use when deploying LLMs and needing real-time safety monitoring, calibrating safety thresholds, or designing deployment-time guardrails. Simpler than sequential hypothesis testing approaches. Activation: LLM safety monitoring, deployment safety, real-time alarm, verifier thresholding, ri...Votes: 0GitHub stars: 3
- Onnes Llm Cryogenic DiagnosisPhysics-grounded digital twin + multi-agent LLM methodology for fault diagnosis in critical infrastructure. Uses a forward physics model with learned noise fingerprint to drive LLM agents for diagnostic tasks. Trigger words: fault diagnosis, digital twin, multi-agent LLM, cryogenic, physics-grounded, diagnostic simulator.Votes: 0GitHub stars: 3
- Onnes Physics Grounded Llm Digital TwinOnnes methodology — physics-grounded digital twin simulator driving multi-agent LLM operations layer for cryogenic fault diagnosis. Combines forward physics model with learned noise fingerprint, enabling zero-shot fault classification via contrastive few-shot demonstrations and self-consistency voting.Votes: 0GitHub stars: 3
- Ontoextend Framework Requirement Driven Scalable Ontology Extension LlmsSkill derived from arXiv:2607.17963 - OntoExtend: A Framework for Requirement-driven and Scalable Ontology Extension with LLMsVotes: 0GitHub stars: 3
- Ontology Constrained Llm Hypothesis Scoring本体约束多 LLM 假设评分方法论。使用专家本体（36 个概念）约束本地多 LLM 理事会，对跨学科文献（如预测编码神经科学）进行假设支持评分，生成可审计的分歧测量和定量假设空间映射。Votes: 0GitHub stars: 3
- Openai And Los Alamos National Laboratory AnnounceSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Baselines Acktr A2cSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Baselines DqnSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Five Benchmark ResultsSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Five Defeats Dota 2 World ChampionsSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai FiveSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Gpt 5 System CardGPT-5 system architecture and safety methodology - unified model with reasoning router, safe-completions, and deliberative alignmentVotes: 0GitHub stars: 3
- Openai Gym BetaSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai MicroscopeSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai O1 MiniSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai O1 System Card External Testers AcknowledgeSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai O1 System Cardo1 chain-of-thought reasoning and deliberative alignment methodology - RL-based reasoning training for improved safetyVotes: 0GitHub stars: 3
- Openai O3 Mini System CardSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai O3 MiniSkill for AI agent capabilitiesVotes: 0GitHub stars: 3
- Openai Privacy FilterMethodology for building PII detection and redaction models using bidirectional token-classification with span decoding and constrained Viterbi. Based on OpenAI Privacy Filter (1.5B params, 50M active, 128K context). Use when implementing privacy-preserving NLP pipelines, PII detection systems, or data redaction workflows.Votes: 0GitHub stars: 3