Most Quoted Sentences
The most-quoted sentences across the web, all-time
Updated hourly from live whisper data.
Whispers
- 1
3 Whispers
models the log-odds of an event as a linear combination of one or more independent variables
- 2
3 Whispers
RL is the study of agents and how they learn by trial and error.
- 3
3 Whispers
Kafka combines three key capabilities so you can implement your use cases for event streaming end-to-end
- 4
3 Whispers
models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts
- 5
3 Whispers
Plain C/C++ implementation without any dependencies
- 6
Relying solely on a system prompt crafted with instructions to be careful of injection attempts has limited effectiveness.
- 7
3 Whispers
In Go, a string is in effect a read-only slice of bytes.
- 8
3 Whispers
JSX is a syntax extension for JavaScript that lets you write HTML-like markup inside a JavaScript file.
- 9
Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale.
- 10
Mixtral 8x7B, a Sparse Mixture of Experts (SMoE) language model
- 11
Neural networks are powerful and flexible models that work well for many difficult learning tasks
- 12
overfitting is the production of an analysis that corresponds too closely or exactly to a particular set of data
- 13
Prompt engineering is the process of structuring natural language inputs (known as prompts) to produce specified outputs from a generative artificial intelligence (GenAI) model.
- 14
reinforcement learning is particularly well-suited to problems that include a long-term versus short-term reward trade-off.
- 15
Gradient boosting is a machine learning technique based on boosting in a functional space, where the target is pseudo-residuals instead of residuals as in traditional…
- 16
In a well-dimensioned hash table, the average time complexity for each lookup is independent of the number of elements stored
- 17
Indoor localization services are a crucial aspect for the realization of smart cyber-physical systems within cities of the future.
- 18
Its relative insensitivity to gap length is its advantage over other RNNs
- 19
diffusion models, also known as diffusion-based generative models or score-based generative models, are a class of latent variable generative models
- 20
Keras 3 is a multi-backend deep learning framework, with support for JAX, TensorFlow, PyTorch, and OpenVINO
- 21
Language models are useful for a variety of tasks
- 22
Machine translation is the use of computational techniques to translate text or speech from one language to another
- 23
Generative artificial intelligence (GenAI) is a subfield of artificial intelligence (AI) that uses generative models to generate text, images, videos, audio, software code or other…
- 24
Natural language processing (NLP) is the processing of natural language information by a computer.
- 25
3 Whispers
nginx has one master process and several worker processes.
- 26
Orca: Progressive Learning from Complex Explanation Traces of GPT-4
- 27
process of transferring knowledge from a large model to a smaller one
- 28
3 Whispers
Promise.race is always asynchronous: it never settles synchronously, even when the iterable is empty.
- 29
puts elements of a list into an order
- 30
regularization is a process that converts the answer to a problem to a simpler one.
- 31
Decision trees are among the most popular machine learning algorithms given their intelligibility and simplicity
- 32
generative artificial intelligence chatbot and virtual assistant developed by Google
- 33
How can we perform efficient inference and learning in directed probabilistic models, in the presence of continuous latent variables with intractable posterior distributions, and large…
- 34
If they find a parrot who could answer to everything, I would claim it to be an intelligent being without hesitation.
- 35
3 Whispers
convolution becomes quite a straightforward idea
- 36
in-context learning (ICL) has emerged as a new paradigm
- 37
Instruction tuning large language models (LLMs) using machine-generated instruction-following data has improved zero-shot capabilities on new tasks
- 38
3 Whispers
It can be accessed using:const path = require('node:path')
- 39
decoding K tokens takes K serial runs of the model
- 40
Demystifying Chains, Trees, and Graphs of Thoughts
- 41
3 Whispers
Diffusion models are inspired by non-equilibrium thermodynamics.
- 42
designed for processing sequential data, such as text, speech, and time series
- 43
effect of the convolutional network depth on its accuracy
- 44
dimensionality reduction, thoughtfully applied, can give us a lot of traction on understanding neural networks.
- 45
Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning
- 46
loss scales as a power-law with model size, dataset size, and the amount of compute
- 47
emergence occurs when a complex entity has properties or behaviors that its components do not have on their own
- 48
extends the context window sizes of RoPE-based pretrained LLMs such as LLaMA models to up to 32768
- 49
3 Whispers
Multi-stage builds let you reduce the size of your final image
- 50
3 Whispers
Native ES imports do not support bare module imports
- 51
3 Whispers
Next.js is a React framework for building full-stack web applications.
- 52
Event handlers are your own functions that will be triggered in response to interactions
- 53
normalize the activities of the neurons
- 54
optimal lower bound $\frac{\lambda_n}{\lambda_m}\geq(\frac{n}{m})^2$ of vibrating string
- 55
3 Whispers
FastAPI is a modern, fast (high-performance), web framework for building APIs with Python based on standard Python type hints.
- 56
generative adversarial network (GAN) is a class of machine learning frameworks
- 57
first-order iterative algorithm for minimizing a differentiable multivariate function
- 58
Finetuning language models on a collection of datasets phrased as instructions has been shown to improve model performance and generalization to unseen tasks.
- 59
3 Whispers
For example
- 60
Props are the information that you pass to a JSX tag.
- 61
Random forests are a way of averaging multiple deep decision trees
- 62
Redundant or duplicate state is a common source of bugs.
- 63
a new Python toolkit called RecWizard for Conversational Recommender Systems
- 64
CNNs are also known as shift invariant or space invariant artificial neural networks
- 65
3 Whispers
Classes provide a means of bundling data and functionality together.
- 66
Effects let you specify side effects that are caused by rendering itself, rather than by a particular event.
- 67
Graph neural networks (GNNs) are artificial neural networks designed for tasks whose inputs are graphs.
- 68
Hermann Bottenbruch published the first implementation to leave out this check in 1962
- 69
3 Whispers
hyperlinking is good RESTful design
- 70
Hyperparameter optimization determines the set of hyperparameters that yields an optimal model
- 71
3 Whispers
Constructs a closure (an unnamed function object capable of capturing variables in scope).
- 72
3 Whispers
Constraints give you as much control over the data in your tables as you wish.
- 73
3 Whispers
In MongoDB
- 74
In React, data that changes over time is called state.
- 75
A typical generative task is as follows. At each step, a datapoint is sampled from the dataset, and part of the data is removed, and…
- 76
a word embedding is a representation of a word
- 77
3 Whispers
Interpretability techniques are normally studied in isolation.
- 78
activation function of a node is a function that calculates the output of the node based on its individual inputs and their weights
- 79
Aggregate functions compute a single result from a set of input values.
- 80
3 Whispers
AI is the field devoted to building artifacts that are intelligent
- 81
AI safety is an interdisciplinary field focused on preventing accidents, misuse, or other harmful consequences arising from artificial intelligence systems.
- 82
deep learning studies a connection between optimization and functional programming.
- 83
All modern browsers support module features natively without needing transpilation.
- 84
denoising autoencoder for pretraining sequence-to-sequence models
- 85
3 Whispers
Always use string, number, or boolean for types.
- 86
an algorithm for first-order gradient-based optimization of stochastic objective functions
- 87
distribution of each layer's inputs changes during training
- 88
An autoencoder is a type of artificial neural network used to learn efficient codings of unlabeled data
- 89
An empty arrow function returns undefined
- 90
Deeper neural networks are more difficult to train.
- 91
Large language models (LLMs) have achieved impressive performance on code generation.
- 92
Large Language Models are Zero-Shot Reasoners
- 93
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- 94
During development, Vite assumes that a modern browser is used.
- 95
any model that is trained on broad data
- 96
3 Whispers
Any variable may be used as an indexed array
- 97
Efficient management of attention key and value memory with PagedAttention
- 98
Arrays are mutable in JavaScript, but you should treat them as immutable when you store them in state.
- 99
Momentum is a heavy ball rolling down the same hill.
- 100
3 Whispers
Most machine learning workflows involve working with data