All authors
Claude Skills by ADu2021
github.com/ADu20211,228 skills0 installs1,664 views
- Mixture Of Depths AttentionAllow attention heads to reference features from multiple depths by accessing both current-layer and depth key-value pairs. Prevent signal degradation in deep models while maintaining computational efficiency.Votes: 0GitHub stars: 6
- Mixture Of Experts Adaptive CapacityOptimize Mixture-of-Experts efficiency by decoupling token-level expert activation from layer architecture. Use dynamic threshold routing where expert count per token varies by input complexity, and apply layer-wise capacity scheduling to match representational diversity patterns.Votes: 0GitHub stars: 6
- Mixture Of Reasonings Adaptive StrategiesEnable LLMs to autonomously select and apply diverse reasoning strategies without prompt engineering. Trains models with diverse thought templates covering 50-500 distinct reasoning approaches, achieving 2-13% improvements over baseline prompting methods.Votes: 0GitHub stars: 6
- Mixture Of Recursions Adaptive ComputationBuild parameter-efficient models that assign different computation depths per token via learned routing, combining weight sharing with dynamic computation complexity. Use when you need to maximize model capacity within compute budgets or create models that allocate compute adaptively based on token complexity.Votes: 0GitHub stars: 6
- Mm Helix ReasoningTrain multimodal models for long-chain reflective reasoning (iterative thinking, backtracking) using Adaptive Hybrid Policy Optimization. Trigger: improve VLM performance on tasks requiring iterative refinement and error correction.Votes: 0GitHub stars: 6
- Mmdeepresearch Bench A Benchmark For MultimodalDeep Research Agents (DRAs) generate citation-rich reports via multi-step search and synthesis, yet existing benchmarks mainly target text-only settings or short-form multimodal QA, missing end-to-end multimodal evidence use. We introduce MMDeepResearch-Bench (MMDR-Bench), a benchmark of 140 expert-crafted tasks across 21 domains, where each task provides an image-text bundle to evaluate multimodal understanding and citation-grounded report generation. Compared to prior setups, MMDR-Bench emp...Votes: 0GitHub stars: 6
- Mmgr Multimodal ReasoningEvaluate whether generative models possess foundational reasoning capabilities. Develop five-ability framework (physical, logical, 3D spatial, 2D spatial, temporal reasoning) across abstract reasoning, embodied navigation, and physical commonsense benchmarks. Use structured rubric requiring simultaneous satisfaction of all sub-metrics.Votes: 0GitHub stars: 6
- Mobe Mixture Basis ExpertsCompresses MoE language models through shared basis factorization of expert weight matrices, achieving 24-30% parameter reduction with minimal accuracy loss.Votes: 0GitHub stars: 6
- Mobile Agent V3 Gui AutomationBuild GUI automation agents using self-evolving trajectory generation, trajectory-aware policy optimization, and integrated action semantics for cross-platform interaction.Votes: 0GitHub stars: 6
- Moca Multimodal EmbeddingsTransform pre-trained vision-language models into powerful bidirectional multimodal embeddings through modality-aware continual pre-training and heterogeneous contrastive fine-tuning. 3B model matches 7B baselines.Votes: 0GitHub stars: 6
- Modality Adaptive Reasoning VisualizationsEnable small vision-language models to reason over diverse data types by converting latent embeddings into visual representations, achieving specialized performance without domain-specific training.Votes: 0GitHub stars: 6
- Model Merging Dual AnchorsMerges multiple fine-tuned models by operating in input-representation space rather than parameter space. Creates synthetic inputs whose gradients align with task vectors, bridging joint training and post-hoc merging for robust multi-task model combination.Votes: 0GitHub stars: 6
- Modomodо Multimodal RlOptimize data mixtures across diverse vision-language domains when applying RL with verifiable rewards to multimodal LLMs, balancing task-specific performance with generalization.Votes: 0GitHub stars: 6
- Modular Large Model TrainingTrain large models efficiently across heterogeneous hardware (GPUs, TPUs, Trainium) using strict encapsulation principles, achieving constant code complexity when adding features across hundreds of modules.Votes: 0GitHub stars: 6
- Moe Routing AlignmentPrevents MoE router instability during RL training by recording and replaying inference-phase routing distributions back into training. Reduces training-inference routing divergence and KL divergence, enabling stable MoE RL scaling without sacrificing training speed.Votes: 0GitHub stars: 6
- Moe Sparsity ReasoningDetermine optimal MoE sparsity by separating memorization and reasoning trade-offs: active FLOPs improve reasoning while total parameters improve memorization, requiring joint optimizationVotes: 0GitHub stars: 6
- Molecular Thought ReasoningImprove agent reasoning by designing thought structures that balance deep analysis, self-reflection, and exploratory thinking. Framework discovers that effective long-form reasoning exhibits molecular-like interaction patterns—specific bonds between reasoning components that enable fast entropy convergence. Method synthesizes improved reasoning trajectories using distribution-transfer, improving both model performance and RL training stability.Votes: 0GitHub stars: 6
- Monet Latent Visual ReasoningEnable multimodal LLMs to generate and reason with latent visual embeddings as intermediate thoughts: implement supervised fine-tuning to produce continuous visual representations, then optimize via VLPO to treat embeddings as learnable actions in reinforcement learning.Votes: 0GitHub stars: 6
- Moss Transcribe Diarize Accurate Transcription WitResearch contribution advancing agent and reasoning capabilities through novel approaches to model development, training, and evaluation.Votes: 0GitHub stars: 6
- Motion Stream Real Time Interactive VideoGenerate videos at 29 FPS with interactive motion control through teacher-student distillation of motion-conditioned video models, using sliding-window causal attention and attention sinks to maintain constant latency for indefinite-length generation.Votes: 0GitHub stars: 6
- Mr Align Meta Reasoning FactualityImprove factuality in large reasoning models by analyzing reasoning state transitions and reweighting preference optimization signals, suppressing defective reasoning segments while amplifying patterns that lead to factual outputs.Votes: 0GitHub stars: 6
- Mr Search Meta Rl Agentic SearchImprove agent search through meta-RL: generate multiple episodes sequentially, each building on prior attempts with explicit self-reflection. Use turn-level RLOO advantage estimation to provide dense credit without value models.Votes: 0GitHub stars: 6
- Msign Stable Rank RestorationPrevent unrecoverable gradient explosions in LLM training by periodically restoring weight matrix stable rank through SVD-based matrix sign operations, eliminating sudden training failures without computational burden.Votes: 0GitHub stars: 6
- Mu Parametrization MoeApply μ-parametrization to Mixture-of-Experts architectures to enable reliable hyperparameter transfer across model sizes, eliminating costly retuning when scaling to trillion-parameter systems.Votes: 0GitHub stars: 6
- Multi Agent EvolveEnables LLM self-improvement without external verification through multi-agent co-evolution. Proposer generates questions, Solver attempts solutions, Judge evaluates both. All agents evolve together via RL, achieving 4.54% improvement on reasoning benchmarks without human supervision.Votes: 0GitHub stars: 6
- Multi Agent Memory FrameworkDesign multi-agent systems with brain-inspired memory mechanisms that enable efficient information sharing and coordination. Implement hierarchical memory structures (working memory, episodic memory, semantic memory) similar to neuroscience models to improve multi-agent reasoning, planning, and task completion.Votes: 0GitHub stars: 6
- Multi Agent Memory SystemBuild persistent, structured memory systems for LLM agents that remember user context across sessions, organize information semantically, and retrieve relevant knowledge automatically before responding. Achieves 35% accuracy gains over RAG baselines with 99.9% smaller storage overhead.Votes: 0GitHub stars: 6
- Multi Agent Tool Policy OptTrain planner and worker agent roles within a single LLM via role-specific prompts and RL, avoiding multi-instance overhead while preserving specialization. Trigger: improve tool-use planning robustness to noisy outputs without deploying separate models.Votes: 0GitHub stars: 6
- Multi Scale Speculative DecodingAccelerate autoregressive image generation via multi-resolution drafting with spatially-informed verification. Local rejection and resampling enable efficient error correction focusing on spatial neighborhoods, achieving 1.7× speedup over baselines.Votes: 0GitHub stars: 6
- Multi Task Grpo RobustEnable balanced multi-task GRPO training via robustness-aware optimization and improvement-aware task reweighting, dynamically adjusting task weights based on both reward and loss trajectory improvement, achieving 6-28% worst-task improvements while maintaining competitive average accuracy.Votes: 0GitHub stars: 6
- Multiagent CommunicationEnable agents to communicate through shared latent thoughts rather than natural language, recovering both shared and private latent representations with theoretical guarantees for more efficient collaboration.Votes: 0GitHub stars: 6
- Multiagent Process RewardsTrain specialized agents in pipelines using dense per-action process rewards from AI coaching. Solves credit assignment in sequential workflows, enabling better generalization and faster convergence than outcome-only training.Votes: 0GitHub stars: 6
- Multimodal Diffusion AlignmentImprove text-image alignment in diffusion transformers through Temperature-Adjusted Cross-modal Attention (TACA), addressing token imbalance and timestep-dependent weighting with parameter-efficient LoRA fine-tuning.Votes: 0GitHub stars: 6
- Multimodal Video Document EmbeddingsGenerate unified embeddings for videos, images, and visual documents enabling semantic similarity, retrieval, and clustering across heterogeneous visual content types.Votes: 0GitHub stars: 6
- Multiverse Parallel GenerationEnable native parallel token generation in language models by implementing adaptive task decomposition and merge strategies, achieving 2x speedup with 1.87% performance gains.Votes: 0GitHub stars: 6
- Muon Optimizer Tail Memory LearningImprove LLM training efficiency by selectively applying Muon optimizer to Value-Output attention weights and FFN layers, which function as associative memories. Use when training data exhibits heavy-tailed distributions requiring robust rare-fact learning.Votes: 0GitHub stars: 6
- Muses Designing Composing Generating Nonexistent FEnhanced language model pre-training methodology improving linguistic competence across languages, strengthening foundational capabilities for multilingual agent systems.Votes: 0GitHub stars: 6
- Musixqa Visual Music UnderstandingTeach multimodal LLMs to read and understand sheet music through a synthetic QA dataset with kern+ symbolic notation. Enables models to handle music sheet OCR, symbol recognition, and chord estimation at 8× better performance than GPT-4o.Votes: 0GitHub stars: 6
- Nabla ReasonerImproves LLM reasoning quality at inference time by optimizing token logits using gradient descent, combining reward model signals with KL-regularization. Bridges parametric training-time and non-parametric test-time scaling through token-level optimization.Votes: 0GitHub stars: 6
- Nag Diffusion GuidanceApply training-free negative guidance in diffusion models by extrapolating in attention space with L1-based normalization, restoring suppression of unwanted attributes across architectures and modalities.Votes: 0GitHub stars: 6
- Native Parallel ReasonerEnable LLMs to develop genuine parallel reasoning without external supervision through progressive self-distilled training. Transform models from sequential reasoning to native parallel cognition with 4.6× speedup—ideal when latency and reasoning quality both matter.Votes: 0GitHub stars: 6
- Nbdiff Block Diffusion LlmConvert auto-regressive language models to efficient diffusion-based generators through gradual block size increments. NBDiff-7B inherits long-context capabilities from AR predecessors while achieving state-of-the-art parallel generation—ideal when you need efficiency without sacrificing reasoning.Votes: 0GitHub stars: 6
- Nemotron 3 NanoEfficient agentic reasoning via sparse MoE activating 50% parameters per token. Combines Mamba-Transformer hybrid with 6-of-128 expert routing, three-stage post-training (SFT, verifiable RL, RLHF), and Group Relative Length Control—achieving 3.3× inference throughput of competitors while maintaining 1M token context support and superior reasoning.Votes: 0GitHub stars: 6
- Nemotron 3Build efficient open-source LLMs via hybrid Mamba-Transformer MoE architecture with LatentMoE expert design, multi-token prediction training, FP4 precision, and multi-environment RL post-training—achieving 3.3× higher throughput than equivalently-sized models while maintaining state-of-the-art reasoning, coding, and tool-use capabilities.Votes: 0GitHub stars: 6
- Nemotron Cascade RlTrain language models through sequential, domain-wise RL stages (RLHF → Instruction-Following → Math → Code → SWE) without catastrophic forgetting. Exploit policy-dependent training data distribution where previous behaviors persist when reward-relevant. 14B model surpasses DeepSeek-R1-0528 (671B) on LiveCodeBench.Votes: 0GitHub stars: 6
- Nemotron Elastic Efficient ReasoningDeploy multiple reasoning model sizes efficiently by embedding nested submodels within a single parent—use end-to-end trained routers to select submodels at inference, achieving 360× cost reduction vs training families separately.Votes: 0GitHub stars: 6
- Nemotron Flash Latency OptimalEvolutionary architecture search discovering optimal depth-width ratios and operator combinations under deployment latency constraints using augmented scaling laws. Deploy when you need fastest inference per latency target with mixed attention mechanisms.Votes: 0GitHub stars: 6
- Nemotron Math Long ContextCreate 7.5M long-form mathematical solution traces with multi-mode supervision (high/medium/low reasoning depths, with/without Python). Integrate 85K competition problems (AoPS) and 262K community questions (StackExchange). Implement sequential bucketing training achieving 2-3× speedup while maintaining accuracy.Votes: 0GitHub stars: 6
- Nemotron Nano Hybrid Mamba TransformerBuild hybrid Mamba-Transformer models combining efficient Mamba-2 layers with standard attention to achieve 6x higher inference throughput while maintaining reasoning accuracy on long-context tasks.Votes: 0GitHub stars: 6
- Nepa Next Embedding PredictionTrain vision transformers through autoregressive next-embedding prediction without pixel reconstruction, tokenizers, or contrastive losses. Apply causal masking and stop-gradient on target embeddings. Achieve 83.8% (ViT-B) and 85.3% (ViT-L) ImageNet-1K accuracy with strong transfer to downstream tasks.Votes: 0GitHub stars: 6