3,080 technical terms and definitions
self-assembly in nanostructures, self-assembly nanostructures
Self-Assembly in Nanostructures Introduction & Motivation Self-Assembly in Nanostructures addresses a central problem in self-assembly process control and assembly-error-rate…
scaled dot product attention, query key value attention, transformer self-attention, multi-head attention
Self-attention lets every token in a sequence look at every other token and decide, for itself, which ones matter most to its own meaning right now — each token projects itself…
audio & speech
Self-Attention ASR is speech recognition architectures that rely heavily on transformer self-attention encoders or decoders - They model long-range dependencies in speech more…
Self-Attention GAN Introduction & Motivation Self-Attention GAN addresses a central problem in generative generation and long-range-consistency synthesis: how to Design generative…
neural architecture
Self-Attention in Capsules is the architectural innovation that replaces the original slow iterative routing algorithm in Capsule Networks with the parallelizable self-attention…
self-attention query-key-value projection and similarity prediction, self-attention query-key-value projection, query-key-value projection similarity, query-key-value projection
Self-Attention Query-Key-Value Projection and Similarity Prediction Introduction & Motivation Self-Attention Query-Key-Value Projection and Similarity Prediction addresses…
self attention scaled dot product, self attention, scaled dot product, bert bleu qk, self-attention scaled dot-product, self-attention scaled, scaled dot-product
Self-Attention - Scaled Dot-Product Introduction & Motivation Self-attention: relate sequence elements to each other. Scaled dot-product attention mechanism. Applications…
time series models
Self-attentive Hawkes is a Hawkes-style event model augmented with self-attention to represent nonlocal event influence - Self-attention weights identify which historical events…
self-checking testbench
Self-Checking Testbench Introduction & Motivation Self-Checking Testbench addresses a central problem in verification process control and assertion-density-per-cycle prediction…
majority vote
Self-Consistency Decoding What is Self-Consistency? Self-consistency generates multiple reasoning paths for the same problem, then selects the most common final answer through…
reasoning
Self-consistency improves reasoning accuracy by generating multiple solution paths and selecting the most common answer. Mechanism: Sample N reasoning chains with temperature > 0…
prompting
Self-consistency is the reasoning strategy that samples multiple independent solution paths and selects the most frequent final answer - it improves robustness by aggregating over…
prompting techniques
Self-Consistency is a reasoning strategy that samples multiple solution paths and selects the most consistent final answer - It is a core method in modern LLM workflow execution.…
prompting
Self-critique is the prompting approach where a model evaluates weaknesses in its own answer using explicit review criteria - it separates generation from quality assessment…
prompting techniques
Self-Critique is a prompting approach that asks the model to evaluate weaknesses in its own answer before finalizing - It is a core method in modern LLM workflow execution. What…
ai safety
Self-Critiquing is an AI safety technique where the model evaluates and critiques its own outputs — generating initial responses, then assessing them for errors, harmfulness…
consistency, regularize, augmentation, born-again
Self-distillation trains a model to match its own predictions on augmented or different views of data — using the model itself as both teacher and student to improve consistency…
Self-Distillation Introduction & Motivation Self-Distillation addresses a central problem in reducing the size and cost of compression models: how to Reduce the size and inference…
model compression
Self-Distillation is a knowledge distillation technique where the teacher and student share the same architecture — the model distills knowledge into itself, either by using…
self-supervised learning
Self-distillation is the training strategy where a model learns from softened targets produced by another instance of itself or an exponential moving average teacher - in vision…
model optimization
Self-Distillation is a method where a model learns from its own earlier states or auxiliary heads - It improves performance without requiring a separate external teacher model.…
Self-Distillation (Extended Study) Introduction & Motivation Self-Distillation (Extended Study) addresses a central problem in reducing the size and cost of compression models…
Self-Distillation (Extended Study) Introduction & Motivation Self-Distillation (Extended Study) addresses a central problem in reducing the size and cost of compression models…
Self-Distillation (Extended Study) Introduction & Motivation Self-Distillation (Extended Study) addresses a central problem in reducing the size and cost of compression models…
domain adaptation
Self-Ensembling for Domain Adaptation refers to domain adaptation methods that use temporal ensembling or mean teacher techniques—where a slowly-updated copy of the model…
evaluation
Self-evaluation in AI refers to a model's ability to assess, critique, and score its own outputs. This metacognitive capability enables language models to identify errors, rate…
senn, explainable ai
SENN (Self-Explaining Neural Networks) are neural networks architecturally designed to produce their own explanations alongside predictions — generating interpretable concept…
Self-Explaining Neural Networks SENN with Concepts Introduction & Motivation Self-Explaining Neural Networks SENN with Concepts addresses a central problem in explainability…
neural architecture
Self-Gating is a mechanism where a neural network layer gates its own activations using a function of the same input — the input multiplied by a sigmoid (or similar gate)…
finfet self heating, local channel temperature, electro thermal simulation, fin thermal isolation
Self-Heating in FinFETs Self-heating is the temperature rise inside a transistor's own active channel caused by the power it dissipates while switching, trapped locally because…
reliability
Silicon-on-Insulator (SOI) substrate engineering, Fully Depleted SOI (FD-SOI) planar architectures, and dynamic back-gate body biasing constitute the engineered substrate…
signal & power integrity
Self-Heating Interconnect is temperature rise of interconnect structures caused by their own current-induced dissipation - It can create localized hotspots independent of nearby…
simulation
Self-heating modeling is the electrothermal modeling of temperature rise generated internally by device operation and limited heat extraction - it predicts local channel…
on prem, local deploy
Self-Hosting LLMs is the deployment of large language models on your own infrastructure (on-premise servers, private cloud, or dedicated GPU instances) rather than using…
quality & reliability
Self-Inspection is operator-performed verification of completed work against clear criteria before transfer - It is a core method in modern semiconductor quality engineering…
data generation
Self-Instruct is a data-generation pipeline where a model creates synthetic instructions and responses for further tuning - Bootstrapped generation expands instruction coverage…
training techniques
Self-Instruct is a data-generation method where models synthesize instruction-output examples to bootstrap instruction tuning - It is a core method in modern LLM training…
self-play, TD-Gammon, Tesauro 1992, AlphaZero, reinforcement learning, temporal difference learning, learning from experience
A self-learning program is software that improves its own competence from experience it generates or collects itself, rather than from a fixed rule set, and the clearest example…
ai agents
Self-Monitoring is continuous tracking of internal agent state to detect loop, drift, or instability conditions - It is a core method in modern semiconductor AI-agent coordination…
Self-Paced Active Learning SPAL with Hardness Regularizer Introduction & Motivation Self-Paced Active Learning SPAL with Hardness Regularizer addresses a central problem…
machine learning
Self-Paced Learning is a curriculum learning variant where the model itself decides which training examples to include — the model's own loss on each example determines…
advanced training
Self-paced learning is a learning approach where models select training samples based on current confidence and difficulty - The model starts with high-confidence examples…
self-paced learning sample weighting and adaptive sampling, self-paced learning sample, learning sample weighting, sample weighting adaptive, weighting adaptive sampling
Self-Paced Learning Sample Weighting and Adaptive Sampling Introduction & Motivation Self-Paced Learning Sample Weighting and Adaptive Sampling addresses a central problem…
Self-Play and Competitive Learning Introduction & Motivation Self-Play and Competitive Learning addresses a central problem in reinforcement policy learning…
alphago, alphazero, self play training, game play ai
Self-Play Reinforcement Learning is the training paradigm where an AI agent improves by playing against copies of itself — generating its own training data through…
rag
Self-RAG enables models to decide when retrieval is needed versus generating from internal knowledge. Motivation: Not every query needs retrieval - simple questions answered…
rag
Self-RAG is the retrieval-augmented generation approach where the model learns to reflect on answer quality and decide when to retrieve additional evidence - it integrates…
rag
Self-RAG is a reflective RAG approach where the model decides when to retrieve, evaluate context quality, and revise outputs - It is a core method in modern RAG and retrieval…
prompting
Self-refine is the iterative prompting method where a model repeatedly generates output, evaluates it, and refines it toward better quality - it formalizes draft-to-revision…
prompting techniques
Self-Refine is an iterative generation loop that alternates draft creation, critique, and revision to improve output quality - It is a core method in modern LLM workflow…