1,517 technical terms and definitions
ai tool use, function calling, llm tools, api calling, structured tool call, react agent
Tool-use AI allows a model to request structured calls to external functions, APIs, search, databases, code runtimes, devices, or business systems. Tools extend a model beyond…
ATE test socket, DUT socket, pogo pin contactor, clamshell socket, final test interface, socket insertion cycle life, contact resistance test socket
A test contactor is the mechanical interposer that makes temporary, repeatable electrical contact between a packaged die's leads or balls and the ATE load board beneath it, so…
dopant fluctuation device mismatch, work function metal gate, fin-fet undoped channel, sram cell variability scaling
Channel doping is the controlled introduction of dopant atoms into the transistor channel region — to adjust threshold voltage (V_t), control short-channel effects, and modulate…
cmos invention 1963 step 14 wanlass named inventor sah co author contested framing, cmos invention priority dispute
Trace the Priority Question: Who Gets Called the Inventor, and Why It Is Contested 1. How the Record Actually Assigns Credit The patent lists Wanlass as the named inventor…
cmos invention 1963 step 18 citation pattern specialist then general audience retelling, cmos origin story citation history
Trace the Papers Citation Record and How the Story Got Retold Over Time 1. A Rough Shape of the Citation Record Over Time For roughly the first decade after publication, citations…
plate etch dram capacitor, TiN top electrode, SiGe top electrode, capacitor plate contact etch
The top electrode etch is the step that caps off a DRAM cell capacitor by patterning the TiN or SiGe plate layer draped over every storage node cylinder in the array, cutting it…
eprom eeprom history step 2 trace the broken bond wire accident that led to the insight, eprom eeprom history step 2
Trace the Broken Bond-Wire Accident That Led to the Insight 2. An Accident, Followed Up Rather Than Discarded The specific circumstance behind Frohman's insight was mundane…
euv outgassing barrier, mirror contamination resist outgassing, metal oxide resist topcoat free
EUV Resist Topcoat EUV lithography happens entirely inside a vacuum chamber, because EUV light at 13.5nm is absorbed by air and can't travel through it at all. That vacuum…
first commercial mos linear ics step 14 overview, the ua741 what specifically made it commercially dominant
The uA741: What Specifically Made It Commercially Dominant 1. Internal Compensation Removes the Designer's Burden The uA741, introduced in 1968, took the uA709's proven gain…
first commercial mos linear ics step 17 overview, the 78xx series fixed regulators that became a universal cat
The 78xx Series: Fixed Regulators That Became a Universal Catalog Part 1. Simplicity as the Winning Feature Following the adjustable uA723, the fixed-voltage 78xx series…
first commercial mos linear ics step 5 overview, the density for speed trade why this series calls it commerc
The Density-for-Speed Trade: Why This Series Calls It Commercially Decisive, Not Just Technically Real 1. A Tradeoff Already Named, Applied Here Commercially…
pick-and-place accuracy, packaging
Test and packaging turn a completed wafer into a dependable product. Fabrication creates transistors and interconnect, but it does not prove that every die works, connect the die…
power mosfet igbt step 3 power mosfet current path source channel drain, power mosfet current path source channel drain
Trace the Current Path from Source Through Channel to Drain 3. Trace the Current Path from Source Through Channel to Drain Tracing one VDMOS cell's current path end to end shows…
power mosfet igbt step 13 igbt conductivity modulation pnp drift region, igbt conductivity modulation pnp drift region
Trace Conductivity Modulation Through the PNP Output Stage 13. Trace Conductivity Modulation Through the PNP Output Stage The IGBT's improved high-voltage conduction comes…
lithography yield, projection lithography 1973
Trace Micralign's Adoption Across the Industry Chipmakers evaluating Micralign against their existing contact-printing lines weighed the new system's higher throughput, extended…
thin films, thin film growth, film nucleation, nucleation density, growth mode, volmer-weber, frank-van der merwe, stranski-krastanov, percolation threshold, film continuity, structure zone model, resistivity size effect, fuchs-sondheimer, mayadas-shatzkes
A thin film is not merely a thin piece of the bulk material whose name appears on the process traveler. Copper can acquire several times its bulk resistivity, a nominal barrier…
sram six transistor step 4 read operation read disturb, sram read operation read disturb margin
Trace the Read Operation Without Disturbing the Stored State 4. Trace the Read Operation Without Disturbing the Stored State Reading a 6T cell is deceptively delicate: the same…
sram six transistor step 5 write operation overpower feedback, sram write operation overpower latch
Trace the Write Operation: Overpowering the Cell's Own Feedback 5. Trace the Write Operation: Overpowering the Cell's Own Feedback Writing a 6T cell means doing, deliberately…
sram six transistor step 15 multi port register file variants, sram multi port register file cell
Trace Multi-Port and Register-File Variants Built on the Same 6T Core 15. Trace Multi-Port and Register-File Variants Built on the Same 6T Core The 6T cell described in Steps 2…
privacy
T-Closeness is the privacy model requiring that the distribution of sensitive attribute values within each equivalence class be close to the distribution in the overall dataset…
training techniques
T-Closeness is privacy criterion requiring each anonymity group to keep sensitive-value distribution close to the overall population distribution - It is a core method in modern…
t gate, mushroom gate, cutoff frequency ft, cutoff frequency, rf design, ft fmax
Radio frequency (RF), millimeter-wave (mmWave), and sub-terahertz semiconductor transistor architectures constitute the core analog frontend and high-frequency mixed-signal…
T-Learner Two-Model Separated Treatment Estimator Introduction & Motivation T-Learner Two-Model Separated Treatment Estimator addresses a central problem in causal inference…
t sne umap non linear embedding visualization, t sne, umap non linear, 2d 3d kl mnist nn passed, t-sne & umap non-linear embedding & visualization, t-sne umap non-linear, t-sne umap
t-SNE & UMAP: Non-Linear Embedding & Visualization Introduction & Motivation t-SNE: t-Distributed Stochastic Neighbor Embedding. UMAP: Uniform Manifold Approximation…
spc
T-squared chart is the multivariate SPC chart based on Hotelling T-squared statistic to monitor joint deviation from a multivariable process center - it compresses correlated…
quality & reliability
T-Test is a parametric hypothesis test used to compare means between two groups under defined assumptions - It is a core method in modern semiconductor statistical experimentation…
t0, foundation model
T0 is a prompted multitask training framework that fine-tunes models on many natural-language task formulations - T0 uses prompt templates and supervised targets to align model…
t0, training techniques
T0 is a multitask prompted model trained to follow natural-language task instructions across many datasets - It is a core method in modern LLM training and safety execution. What…
generative models
T2I-Adapter is the lightweight adapter module that injects structural conditions into text-to-image diffusion models with low training overhead - it offers controllable generation…
t2t-vit, tokens-to-token vit, computer vision
T2T-ViT (Tokens-to-Token Vision Transformer) is a vision Transformer architecture that introduces a progressive tokenization module that iteratively aggregates neighboring tokens…
T5 Unified Text-to-Text Transformer with Span Corruption Introduction & Motivation T5 Unified Text-to-Text Transformer with Span Corruption addresses a central problem…
t5 text-to-text framework and unified task representation, t5 text-to-text framework, text-to-text framework unified, framework unified task, unified task representation, unified task
T5 Text-to-Text Framework and Unified Task Representation Introduction & Motivation T5 Text-to-Text Framework and Unified Task Representation addresses a central problem in T5…
t5, text-to-text transfer transformer, foundation model
T5 (Text-to-Text Transfer Transformer) is Google's unified NLP model that reframes every language task as a text-to-text problem — both input and output are always text strings…
t5 text to text transfer transformer, t5 text text transfer transformer, t5 text, text transfer transformer, 11b eos glue nlp qa t5, t5 text-to-text transfer, text-to-text transfer transformer
T5 - Text-to-Text Transfer Transformer Introduction & Motivation T5: unified text-to-text framework. Encoder-decoder architecture, diverse NLP tasks. Applications: translation…
taaf, business & standards
TAAF is the test-analyze-and-fix reliability-closure loop used to identify failures, isolate root causes, and implement corrective actions - It is a core method in advanced…
Table Extraction from Documents Introduction & Motivation Table Extraction from Documents addresses a central problem in multimodal learning and table-structure-accuracy…
SQL join types, inner join, left join, full outer join, many-to-many join duplication
A table join is a SQL operation that combines rows from two or more tables based on a related column between them, reconstructing a connected view from data that was deliberately…
nlp
Table-to-text is the NLP task of generating natural language descriptions from tabular data — automatically converting rows, columns, and cells in tables into coherent, accurate…
table to text generation, table to text, to text generation, table to, to text, text generation
Table to Text Generation Introduction & Motivation Table to Text Generation addresses a central problem in nlp process control and table-to-text-structured-data-generation…
Table-to-Text Generation Introduction & Motivation Table-to-Text Generation addresses a central problem in generative generation and table-coverage-accuracy synthesis: how…
completion, local
Tabnine is an AI code completion tool that differentiates itself through local/private model deployment options, enabling enterprises in regulated industries to use AI code…
tabu search diversification, tabu search, search diversification
Tabu Search Diversification Introduction & Motivation Tabu Search Diversification addresses a central problem in ml process control and solution-quality-local-optima prediction…
tabnet, ft transformer, deep learning tables, gradient boosting vs neural
Deep Learning for Tabular Data is the application of neural network architectures specifically designed for structured/tabular datasets — where gradient boosted decision trees…
tabnet feature selection, ft-transformer tabular, entity embedding categorical, gradient boosting vs deep
Deep Learning for Tabular Data is the application of neural networks to tabular/structured data (spreadsheets, databases) — addressing challenges of categorical features, mixed…
VIME Variational Informative Masking for Tabular SSL Introduction & Motivation VIME Variational Informative Masking for Tabular SSL addresses a central problem in selfsupervised…
audio & speech
Tacotron is a neural text-to-speech model that maps text to mel spectrograms with sequence-to-sequence attention - Encoder-decoder attention learns alignment between phonetic…
audio & speech
Tacotron2 is an improved text-to-speech pipeline that combines Tacotron-style spectrogram prediction with neural vocoding - Mel spectrogram generation is paired with high-fidelity…
Tactile-Visual Sensor Fusion for Robotic Manipulation Introduction & Motivation Tactile-Visual Sensor Fusion for Robotic Manipulation addresses a central problem in multimodal…
TADAM Task-Dependent Adaptive Metric Learning Introduction & Motivation TADAM Task-Dependent Adaptive Metric Learning addresses a central problem in fewshot…
tagnn, recommendation systems
TAGNN is a target-aware graph-neural-network recommender that conditions session representation on candidate items - Target-aware attention highlights session nodes most relevant…