← Back to AI Factory Chat

AI Factory Glossary

189 technical terms and definitions

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Showing page 2 of 4 (189 entries)

ternary neural networks,model optimization

Use three values (−1 0 +1).

test case generation from spec, code ai

Generate tests from specifications.

test generation,code ai

Automatically create unit tests for code.

test-time adaptation, domain adaptation

Adapt model during inference without labels.

test-time training, domain adaptation

Update model at test time.

tetrad causal, time series models

Tetrad implements causal discovery algorithms like PC and FCI for learning directed acyclic graphs from data.

text encoder for diffusion, generative models

Encode prompts (CLIP ViT-L).

text-guided image editing, generative models

Edit images using text prompts.

text-to-3d, multimodal ai

Text-to-3D methods generate three-dimensional models from textual descriptions.

text-to-image alignment, generative models

Ensure generated images match text.

text-to-image generation,generative models

Create images from text descriptions (DALL-E Midjourney Stable Diffusion).

text-to-image translation, multimodal ai

Generate images from text.

text-to-sql,code ai

Convert questions to SQL queries.

text-to-video, multimodal ai

Text-to-video models generate video sequences from textual descriptions.

text-to-video,generative models

Generate video clips from text descriptions.

textual inversion, generative models

Learn new tokens from images.

textual inversion, multimodal ai

Textual inversion learns new tokens representing concepts for personalized generation.

textual inversion,generative models

Learn new tokens representing specific concepts for generation.

texture synthesis, multimodal ai

Texture synthesis generates surface appearance through learned or procedural methods.

tgat, tgat, graph neural networks

Temporal Graph Attention Network uses temporal encoding and functional time encoding for dynamic graph learning.

tgcn, tgcn, graph neural networks

Temporal Graph Convolutional Network combines graph convolutions with gated recurrent units to capture spatial and temporal dependencies in dynamic graphs.

tgn, tgn, graph neural networks

Temporal Graph Network maintains memory modules for nodes that are updated during temporal random walks for continuous-time dynamic graph learning.

theory of constraints, supply chain & logistics

Theory of constraints focuses improvement on system bottlenecks maximizing throughput.

theory of constraints, toc, production

Focus on bottleneck.

thermal oxide,diffusion

SiO2 grown by heating silicon in oxygen.

thermal oxidizer, environmental & sustainability

Thermal oxidizers combust VOCs at high temperature converting them to carbon dioxide and water.

thermography maintenance, manufacturing operations

Thermographic inspection detects abnormal temperatures indicating electrical or mechanical problems.

thermoreflectance imaging,failure analysis

Measure temperature via reflectance changes.

threat model,security,design

AI helps threat modeling. Identify attack vectors.

ties-merging, model merging

Merge models by resolving conflicts.

tiled diffusion, generative models

Process image in tiles.

tiling strategy, model optimization

Tiling strategies partition computations into cache-friendly blocks improving memory hierarchy utilization.

time series decomposition, time series models

Time series decomposition separates signals into trend seasonal and residual components for analysis and forecasting.

time-aware attention, graph neural networks

Time-aware attention mechanisms weight edges by recency and temporal patterns in dynamic graphs.

time-based maintenance, production

Maintain on fixed schedule.

time-dependent dielectric breakdown modeling, tddb, reliability

Model oxide breakdown.

time-lagged ccm, time series models

Time-lagged convergent cross mapping identifies optimal delays for detecting causality in nonlinear dynamical systems.

time-resolved emission, failure analysis advanced

Time-resolved photoemission correlates emission timing with circuit activity isolating intermittent or dynamic failures.

timeout agent, ai agents

Timeouts halt agent execution after specified duration preventing runaway processes.

timeout handling, llm optimization

Timeout handling cancels long-running requests preventing resource exhaustion.

timestep embedding, generative models

Encode diffusion timestep.

timm,image models,pretrained

timm (PyTorch Image Models) has hundreds of pretrained vision models. Transfer learning.

tinyml, edge ai

Machine learning on microcontrollers.

tiva (thermally induced voltage alteration),tiva,thermally induced voltage alteration,failure analysis

Defect localization using laser heating.

tiva, tiva, failure analysis advanced

Thermally-Induced Voltage Alteration uses localized heating to modulate pn junction characteristics revealing resistive opens and other defects.

together ai,inference,api

Together AI provides LLM APIs and fine-tuning. Competitive pricing. Open model hosting.

token bucket, llm optimization

Token bucket algorithm allows bursts while maintaining average rate limits.

token budget, llm optimization

Token budgets limit context length for cost and latency management.

token budget,llm architecture

Maximum number of tokens an LLM can process or generate in a single request or conversation turn.

token dropping, llm architecture

Token dropping discards excess tokens when expert capacity is exceeded.