Ying Yao Inventory

LLM-Wiki

Connected notes from Yuque and Notion.

82 notes · 232 relationships
Updated 2026-09-11

82 notes · 232 relationships
RLHF, DPO, and GRPOMHA, MQA, GQA, and MLABackpropagation and Automatic DifferentiationBayesian NetworksBERTCausal AttentionCausal Structure LearningChain-of-Thought ReasoningClassical ML MethodsContext ParallelismContinuous Batching and Sequence PackingData ParallelismDecoding and Sampling StrategiesDeep Learning OptimizationDiffusion ModelsDirect Preference OptimizationDistributed Collective OperationsDistributed Training and MLOps DrillsDistributed Training and MLOps MapDistributed Training and MLOps MapDynamic Sparse AttentionEncoder–Decoder ModelsEngram Conditional MemoryExpert ParallelismFlashAttentionAI and ML Foundations MapAI and ML Foundations MapAI Foundations DrillsFully Sharded Data ParallelismGeneralized Advantage EstimationGenerative Adversarial NetworksGPU Performance Model for LLMsGraph Neural NetworksGreedy Equivalence SearchGroup Relative Policy OptimizationGrouped-Query AttentionKV CacheLLM Compute and Memory AccountingLLM EvaluationManifold-Constrained Hyper-ConnectionsMasked Language ModelingMixed-Precision TrainingMixture of ExpertsML Generalization and RegularizationModel QuantizationMulti-Head Latent AttentionMulti-Query AttentionMulti-Token PredictionNeural Network ComputationNumerical Stability in Neural NetworksPaged AttentionPC AlgorithmPipeline ParallelismPolicy GradientPost-Training and Alignment DrillsProbabilistic ObjectivesPrompt CachingProximal Policy OptimizationRadix AttentionRecurrent Neural NetworksReinforcement LearningResidual NetworksReward ModelingReinforcement Learning from Human FeedbackRoot Mean Square NormalizationRotary Position EmbeddingSearch and A*Self-AttentionSentence EmbeddingsRNNs, Transformers, and State Space ModelsSpectral Residual CNNSpeculative DecodingState Space ModelsSupervised Fine-TuningSwiGLU Feed-Forward NetworkTemporal-Difference LearningTensor ParallelismTime-Series Anomaly DetectionTransformer ArchitectureTransformer Inference MapTransformer Inference MapTransformer Inference DrillsVAE–LOF Anomaly DetectionVariational AutoencoderVector Databases and HNSW
ConceptMapComparisonInterviewSelect a note to read it.