# Language Models and How They Behave

Language models are the component that reads text and predicts what should come next. This area covers how they are built, how they are shaped after training, and the runtime settings that change their behavior on a single call. It also covers the units the whole field is measured in, tokens and context, because nearly every practical limit, cost, and failure you meet later traces back to how text gets split up and how much of it fits at once.

The reason it matters is that most disappointing results are configuration problems rather than capability problems. A model that seems to ignore instructions may be losing them off the end of its context. One that seems creative to a fault may simply be sampling too loosely. Knowing which knob belongs to which layer, training, tuning, prompting, or decoding, turns a vague complaint about quality into a specific question with a specific fix.

## Start here

1. [Token](/en/glossary/token)
2. [Tokenization](/en/glossary/tokenization)
3. [Large Language Model](/en/glossary/large-language-model)
4. [Context Window](/en/glossary/context-window)
5. [Prompt](/en/glossary/prompt)
6. [System Prompt](/en/glossary/system-prompt)
7. [Temperature](/en/glossary/temperature)

## How the pieces fit together

There are four layers, and each one narrows the behavior of the one before it. Pre-training produces a general model from an enormous corpus. Instruction tuning and preference training teach it to follow requests and to favor certain kinds of answers. Prompting shapes a single call, with the system prompt setting standing rules and the user turn carrying the task. Decoding settings then choose among the words the model already considers plausible. Fine-tuning and routing sit alongside these rather than inside them: one changes weights for your domain, the other picks a different model per request without touching any weights at all.

## Where to start

Tokens first, because they are the currency of everything else. Once tokenization is clear, the context window stops being an abstract number and becomes a budget you spend on instructions, history, and retrieved material. From there, treat the prompt and the system prompt as two different instruments rather than one long field. Sampling settings come after that, and they should be the last thing you reach for rather than the first. Training concepts are worth reading once the runtime picture is solid, since they explain why a model responds to instructions in the first place instead of merely continuing text.

## Choosing between models

Treat model choice as a routing decision rather than a single verdict. Different calls in the same product have different needs: a classification step wants speed and consistency, a drafting step wants range, and a step that must emit exact structure wants reliability under constraint. Smaller models handle a surprising share of the work when the task is narrow and the instructions are tight, and open-weight options mostly change where a model can run rather than how it thinks. Build one evaluation set from your own traffic before comparing anything, because that set will disagree with public rankings more often than it agrees.

## What people get wrong

Bigger context is treated as free, and it is not. Filling a window weakens attention to the middle of it and raises both latency and cost, so pruning usually beats stuffing. Benchmarks get read as buying guides, when a leaderboard score says little about your documents and formats. Fine-tuning gets proposed for problems that were really retrieval problems, since a model cannot memorize facts that change weekly. And parameter count gets used as a proxy for quality, which stopped being reliable once data quality, tuning, and sparse architectures started to matter more than raw size.

## Commonly confused

### Pre-training vs Fine-tuning

Pre-training builds general capability from scratch on a huge corpus, while fine-tuning adjusts an already trained model using a far smaller, task-specific set.

### Temperature vs Top-p Sampling

Temperature reshapes how likely each candidate token is, while top-p first trims the candidate list to the most probable ones, so they act at different stages of the same choice.

### Foundation Model vs Large Language Model

Every large language model is a foundation model, but foundation models also include image, audio, and multimodal systems trained in the same broad way.

### Instruction Tuning vs Reinforcement Learning from Human Feedback

Instruction tuning teaches the model to follow the shape of a request, while learning from human feedback teaches it which of two acceptable answers people actually prefer.

## Every term in Language Models and How They Behave

- [Artificial General Intelligence](/en/glossary/artificial-general-intelligence): Artificial general intelligence is a hypothetical form of artificial intelligence able to learn and perform the full range of intellectual tasks that people can perform, rather than excelling only in narrow domains. The term is contested, because researchers disagree about what counts as general capability and about how it would be measured. No agreed definition or accepted test exists.
- [Attention Mechanism](/en/glossary/attention-mechanism): An attention mechanism is a neural network component that lets a model weigh which parts of its input matter most when computing each part of its output. Instead of compressing a whole sequence into one fixed representation, attention computes a set of relevance scores and forms a weighted blend of the input positions. It is the core building block of transformer architectures.
- [Beam Search](/en/glossary/beam-search): Beam search is a decoding strategy that maintains several candidate sequences in parallel, extending each by the most likely next tokens and keeping only the top scoring candidates at every step. The number retained is called the beam width. It finds higher-probability sequences than greedy decoding at proportionally higher compute cost.
- [Benchmark](/en/glossary/benchmark): A benchmark is a standardized set of tasks used to measure and compare model capability, such as graduate-level science questions, competition mathematics, or resolving real software issues. Scores from these sets are widely quoted in model announcements. They are useful for coarse comparison and are weak predictors of performance on any particular real workload.
- [Chain of Thought Prompting](/en/glossary/chain-of-thought-prompting): Chain of thought prompting asks a model to work through intermediate steps before giving a final answer, rather than answering immediately. Producing those steps improves accuracy on arithmetic, logic, and multi-step tasks, because each generated step becomes context for the next. The gain comes from the extra computation the steps allow, not from the model narrating a real internal process.
- [Computer Vision](/en/glossary/computer-vision): Computer vision is the field concerned with extracting information from images and video so that software can describe or act on visual content. Typical tasks include classifying an image, detecting and locating objects, segmenting regions at the pixel level, tracking motion across frames, and reconstructing three dimensional structure. Modern systems are dominated by learned models rather than hand written rules.
- [Constitutional AI](/en/glossary/constitutional-ai): Constitutional AI is a training method in which a model's behavior is shaped by an explicit written set of principles, called a constitution, instead of relying solely on case-by-case human judgments. The model critiques and revises its own responses against those principles, and the revised outputs become training data. The approach makes the values guiding behavior inspectable and editable.
- [Context Window](/en/glossary/context-window): The context window is the maximum number of tokens a model can consider in a single request, counting the prompt, any attached documents, the conversation so far, and the response it generates. Anything outside that budget is simply not available to the model. Limits range from a few thousand tokens in older models to over a million in some current ones.
- [Data Contamination](/en/glossary/data-contamination): Data contamination is the presence of evaluation material in a model's training data, which inflates measured performance because the model has effectively seen the answers. It is widespread with web-scraped corpora, since public benchmarks and their solutions are published online. Contamination makes reported scores unreliable as evidence of generalization.
- [Decoder-Only Model](/en/glossary/decoder-only-model): A decoder-only model is a transformer architecture built from a single stack of causally masked layers that predicts each token from the tokens before it. It has no separate encoder, so input and output occupy the same sequence and the same processing path. Most contemporary text generation systems use this architecture.
- [Deep Learning](/en/glossary/deep-learning): Deep learning is a subfield of machine learning that uses neural networks with many layers of processing units to learn representations directly from raw data. Each layer transforms the output of the previous one, allowing later layers to encode increasingly abstract features. This layered structure underpins most modern speech, image, and language systems.
- [Diffusion Model](/en/glossary/diffusion-model): A diffusion model is a generative system trained to reverse a gradual noising process. During training it learns to remove noise from corrupted data; during generation it starts from pure noise and denoises step by step until a coherent sample emerges. Diffusion is the dominant approach for image, video, and audio generation.
- [Embedding Model](/en/glossary/embedding-model): An embedding model converts text, images, or other data into a fixed-length list of numbers, called a vector, that represents meaning. Items with similar meaning land close together in that numeric space, so similarity can be measured arithmetically. Embeddings underpin semantic search, recommendation, clustering, deduplication, and the retrieval step in retrieval-augmented generation.
- [Emergent Ability](/en/glossary/emergent-ability): An emergent ability is a capability that appears in larger models but is absent or near random in smaller ones of the same family, seemingly arriving abruptly rather than improving gradually with scale. Reported examples include multi-step arithmetic and certain instruction-following behaviors. Whether such jumps are genuine phase transitions or artifacts of how performance is measured is actively disputed.
- [Encoder-Decoder Model](/en/glossary/encoder-decoder-model): An encoder-decoder model is an architecture with two components: an encoder that reads the entire input bidirectionally into a set of representations, and a decoder that generates output while attending to those representations. It was designed for tasks that map one sequence to another, such as translation. It remains common in speech, translation, and some structured transformation systems.
- [Few-shot Prompting](/en/glossary/few-shot-prompting): Few-shot prompting supplies several worked examples of a task inside the prompt so the model can infer the pattern before handling the real input. No parameters change and nothing persists after the request finishes. The technique is also called in-context learning, because the model appears to learn the task from the surrounding context alone.
- [Fine-tuning](/en/glossary/fine-tuning): Fine-tuning continues training an already-trained model on a smaller, targeted dataset so it performs better on a specific task, domain, or format. It changes the model's parameters, which distinguishes it from prompting. Typical uses include enforcing a house style, matching a rigid output schema, or teaching specialized terminology that appears rarely in general training data.
- [Foundation Model](/en/glossary/foundation-model): A foundation model is a large model trained once on broad, general data and then adapted to many different downstream tasks rather than built for a single purpose. Language models, image models, and speech models can all be foundation models. The defining trait is reuse: one expensive training run produces a general base that many separate applications adapt cheaply.
- [Frontier Model](/en/glossary/frontier-model): A frontier model is one of the most capable general-purpose models available at a given time, typically trained at the largest scale currently feasible. The term is used in policy and safety discussion to designate systems whose capabilities are not yet well characterized and may pose novel risks. Its boundary is relative and shifts as the field advances.
- [Generative AI](/en/glossary/generative-ai): Generative AI is a class of machine learning systems that produce new content such as text, images, audio, video, or code, rather than only labeling or scoring data that already exists. These systems learn statistical patterns from large training corpora and then sample fresh outputs consistent with those patterns. The category spans language models, image diffusion models, and speech synthesis systems.
- [Greedy Decoding](/en/glossary/greedy-decoding): Greedy decoding is a text generation strategy that selects the single highest-probability token at every step, with no randomness and no lookahead. It is deterministic, so the same input and model produce the same output every time. It is the simplest decoding method and is equivalent to sampling at a temperature of zero.
- [Inference](/en/glossary/inference): Inference is the act of running a trained model to produce output, as distinct from training, which creates the model. During inference the parameters are frozen and no learning occurs. Every request sent to a deployed model is an inference call, and inference is where the ongoing operational cost of running a model accumulates over its lifetime.
- [Instruction Tuning](/en/glossary/instruction-tuning): Instruction tuning is the training stage that teaches a base model to follow written instructions and answer in a helpful assistant format. It uses a curated set of instruction and response pairs covering many task types. This stage converts a model that merely continues text into one that responds to requests, and it normally precedes preference-based alignment.
- [Intent Classification](/en/glossary/intent-classification): Intent classification is a natural language processing task that assigns an incoming message to one of a predefined set of purposes, such as booking, canceling, or requesting a refund. It converts free text into a routing decision that downstream logic can act on. The label set is designed for a specific application, and unfamiliar requests must be handled explicitly.
- [Knowledge Distillation](/en/glossary/knowledge-distillation): Knowledge distillation is a training technique in which a smaller student model learns to reproduce the behavior of a larger teacher model. The student trains on the teacher's outputs, which may be full probability distributions, generated text, or intermediate representations, rather than only on ground-truth labels. The goal is a compact model that retains much of the teacher's capability.
- [KV Cache](/en/glossary/kv-cache): A KV cache is the stored set of key and value tensors computed for tokens already processed during generation, kept in memory so they do not have to be recomputed for each new token. It converts generation from a repeated full-sequence computation into an incremental one. Its memory footprint grows with sequence length and is a primary limit on serving capacity.
- [Large Language Model](/en/glossary/large-language-model): A large language model is a statistical model trained on very large amounts of text to predict the next piece of text in a sequence. That single objective, applied at scale, produces systems that can answer questions, write and summarize documents, translate, and generate code. The model stores what it learned in billions of numeric weights rather than in a searchable database of documents.
- [Low-Rank Adaptation](/en/glossary/low-rank-adaptation): Low-rank adaptation is a parameter-efficient fine-tuning method that freezes a model's original weights and learns a small pair of matrices whose product forms a correction to selected weight matrices. Because the correction is constrained to a low rank, it holds far fewer trainable values than the weights it modifies. It is the most widely used adapter technique for large models.
- [Machine Learning](/en/glossary/machine-learning): Machine learning is a branch of artificial intelligence in which systems derive rules from data rather than following instructions written by a programmer. A learning algorithm adjusts internal parameters so that a model's outputs better match patterns observed in examples. The resulting model can then make predictions or decisions about data it has not seen before.
- [Max Tokens](/en/glossary/max-tokens): Max tokens is a request parameter that caps how many tokens a model may generate in a single response. Reaching the cap stops generation abruptly, potentially mid-sentence, and the response is reported as truncated by length. It bounds output cost and latency but is distinct from the model's total context window.
- [Mixture of Experts](/en/glossary/mixture-of-experts): Mixture of experts is a model architecture that splits part of the network into many specialized sub-networks, called experts, and routes each token to only a few of them. This lets a model hold a very large total parameter count while activating a small fraction for any given token, giving the capacity of a big model at closer to the compute cost of a smaller one.
- [Model Card](/en/glossary/model-card): A model card is a structured document that describes a model's intended uses, training approach, evaluation results, limitations, and known risks. It accompanies a released model so that adopters can judge fitness for their purpose without reverse-engineering behavior. The format was proposed to make responsible disclosure a routine part of model release.
- [Model Routing](/en/glossary/model-routing): Model routing directs each request to a suitable model instead of sending everything to one. A router may classify difficulty, apply rules by task type, or try a cheap model first and escalate when confidence is low. The goal is to cut cost and latency on easy requests while preserving quality on the hard ones.
- [Multimodal Model](/en/glossary/multimodal-model): A multimodal model accepts or produces more than one type of data, such as text together with images, audio, or video. Inputs from each modality are converted into a shared internal representation so the model can relate them, for example answering a question about a photograph. Most widely used assistants now accept images alongside text and reply in text.
- [Named Entity Recognition](/en/glossary/named-entity-recognition): Named entity recognition is a natural language processing task that locates spans of text referring to specific things and assigns each a type, such as person, organization, location, date, or monetary amount. It converts unstructured text into structured fields that other systems can filter, index, or store. Type schemes are defined per application rather than being universally fixed.
- [Narrow AI](/en/glossary/narrow-ai): Narrow AI, also called weak AI, refers to systems built and trained to perform a specific task or a bounded set of tasks, such as recognizing faces, ranking search results, or transcribing speech. Performance can be very high inside that scope and typically degrades sharply outside it. Every artificial intelligence system in production use today is narrow in this sense.
- [Natural Language Processing](/en/glossary/natural-language-processing): Natural language processing is the field concerned with enabling computers to analyze, interpret, and generate human language. It covers tasks such as translation, summarization, question answering, sentiment analysis, and speech transcription. Approaches have moved from hand written grammatical rules to statistical models and then to large neural models trained on very large text collections.
- [Neural Network](/en/glossary/neural-network): A neural network is a mathematical model made of layers of simple units, each computing a weighted sum of its inputs followed by a nonlinear function. Learning consists of adjusting those weights so that the network's outputs match examples in training data. Networks with many stacked layers are called deep, which is where the term deep learning comes from.
- [Open-weight Model](/en/glossary/open-weight-model): An open-weight model is one whose trained parameters are published for download, so anyone can run it on their own hardware, inspect it, and adapt it. Open weights are not the same thing as open source, since training data and code are usually withheld and licenses often restrict certain uses. The distinction matters legally and is frequently blurred in coverage.
- [Parameter-Efficient Fine-Tuning](/en/glossary/parameter-efficient-fine-tuning): Parameter-efficient fine-tuning is a family of adaptation methods that adjust a small number of parameters while leaving most of a pre-trained model's weights frozen. It achieves much of the benefit of full fine-tuning at a fraction of the memory and storage cost. Common approaches include low-rank adapters, prefix tuning, and bias-only updates.
- [Parameters](/en/glossary/parameters): Parameters are the numeric values inside a model that are adjusted during training and then held fixed when the model is used. They are where everything the model learned is stored. Model size is usually quoted as a parameter count, such as eight billion or several hundred billion, and that number strongly influences memory requirements and running cost.
- [Perplexity](/en/glossary/perplexity): Perplexity is a metric for how well a language model predicts a body of text, computed as the exponential of the average negative log probability the model assigns to each token. Lower values mean the model found the text less surprising. It is the standard intrinsic measure used during pre-training and for comparing models on the same evaluation data.
- [Positional Encoding](/en/glossary/positional-encoding): Positional encoding is the technique of injecting information about token order into a transformer, which otherwise treats its input as an unordered set. The encoding is added to or applied over token representations so that the model can distinguish sequences that contain the same tokens in different arrangements. Several schemes exist, including fixed sinusoidal patterns, learned embeddings, and rotary methods.
- [Pre-training](/en/glossary/pre-training): Pre-training is the first and largest stage of building a model, in which it learns general patterns from a very large unlabeled corpus, typically by predicting the next token. It produces a base model with broad knowledge and language ability but no particular tendency to follow instructions. Later stages then adapt that base into a usable assistant.
- [Prompt](/en/glossary/prompt): A prompt is the complete input given to a model for one request. It usually combines several parts: instructions describing the task, any supplied context or documents, examples, the conversation history, and the user's actual question. The model receives all of this as one sequence of tokens, with no inherent boundary between the parts beyond what its training taught it to expect.
- [Prompt Chaining](/en/glossary/prompt-chaining): Prompt chaining is the practice of decomposing a task into a sequence of separate model calls, where the output of one call becomes part of the input to the next. Each step handles a narrower subtask than the whole. Chaining trades additional calls and latency for higher reliability, easier debugging, and per-step validation.
- [Prompt Engineering](/en/glossary/prompt-engineering): Prompt engineering is the practice of designing and refining the input to a model to get more reliable output. It covers wording instructions precisely, supplying examples, specifying output format, decomposing a task into steps, and testing variants against measured results. It works because a model's behavior is highly sensitive to how a request is framed and structured.
- [Prompt Template](/en/glossary/prompt-template): A prompt template is a reusable prompt structure with placeholders that are filled with variable data at request time. It separates the fixed instructions, formatting rules, and examples from the changing input, so the same tested prompt serves many cases. Templates are the standard way prompts are managed in production systems.
- [Quantization](/en/glossary/quantization): Quantization stores a model's parameters at lower numeric precision, for example using eight-bit or four-bit integers instead of sixteen-bit floating point numbers. This shrinks memory use roughly in proportion to the bit reduction and usually speeds up inference. Accuracy declines with it, typically slightly at eight bits and more noticeably at four bits and below.
- [Reasoning Effort](/en/glossary/reasoning-effort): Reasoning effort is a setting on reasoning-capable models that controls how much internal deliberation a model performs before answering. Higher settings let it generate more hidden reasoning tokens, which improves accuracy on hard problems while increasing latency and cost. Providers expose it either as named levels such as low, medium, and high, or as a numeric token budget.
- [Reinforcement Learning](/en/glossary/reinforcement-learning): Reinforcement learning is a machine learning approach in which an agent learns by interacting with an environment and receiving numerical rewards for the outcomes of its actions. The goal is a policy, meaning a mapping from observed states to actions, that maximizes cumulative reward over time. Learning proceeds through trial, feedback, and revision rather than from labeled examples.
- [Reinforcement Learning From AI Feedback](/en/glossary/reinforcement-learning-from-ai-feedback): Reinforcement learning from AI feedback is a training approach in which preference judgments used to shape a model's behavior come from another model rather than from human annotators. A judging model compares candidate responses against stated principles, and those comparisons train a reward signal or directly optimize the target model. It is used to scale preference training beyond what human labeling can cover.
- [Reinforcement Learning from Human Feedback](/en/glossary/reinforcement-learning-from-human-feedback): Reinforcement learning from human feedback is a training method that shapes a model's behavior using human judgments about which of several candidate responses is better. Those comparisons train a reward model, which then guides further optimization of the language model itself. It is the main technique behind the helpful, cautious tone that most commercial assistants share.
- [Scaling Law](/en/glossary/scaling-law): A scaling law is an empirical relationship describing how a model's loss improves as a power function of model size, training data volume, and compute. These relationships have held across many orders of magnitude, allowing researchers to predict the performance of a large training run from much smaller experiments. They guide how a fixed compute budget is split between parameters and data.
- [Self-Attention](/en/glossary/self-attention): Self-attention is an attention mechanism in which a sequence attends to itself, so every position computes its representation by weighing all other positions in the same sequence. Queries, keys, and values all come from one input rather than from separate sources. It allows a model to build context-sensitive representations of each token based on its surroundings.
- [Sentiment Analysis](/en/glossary/sentiment-analysis): Sentiment analysis is a natural language processing task that assigns an evaluative orientation to text, most often positive, negative, or neutral, and sometimes a numeric intensity or a set of emotion categories. It is applied to reviews, survey responses, support messages, and social posts. Results describe expressed language, which does not always match what a writer actually feels.
- [Small Language Model](/en/glossary/small-language-model): A small language model is a language model with a modest parameter count, often in the range of one to roughly fifteen billion, designed to run cheaply, quickly, and sometimes on a single device. It trades broad general capability for lower latency, lower cost, and easier deployment. The boundary with large models is conventional and shifts as hardware improves.
- [Speculative Decoding](/en/glossary/speculative-decoding): Speculative decoding is an inference optimization in which a small fast model drafts several tokens ahead and a larger target model verifies them in a single parallel pass. Accepted draft tokens are kept and the first rejected one is corrected by the target model. It speeds up generation without changing the output distribution of the target model.
- [Stop Sequence](/en/glossary/stop-sequence): A stop sequence is a string that causes a model to halt generation as soon as it is produced. The matched text is normally excluded from the returned output. Stop sequences give the caller control over where a response ends, independent of the model's own end-of-turn signal or the maximum token limit.
- [Supervised Learning](/en/glossary/supervised-learning): Supervised learning is a machine learning approach in which a model is trained on examples that pair an input with the correct output, called a label. During training the model adjusts its parameters to reduce the difference between its predictions and those labels. Once trained, it assigns outputs to new inputs that were never labeled.
- [Synthetic Data](/en/glossary/synthetic-data): Synthetic data is training data generated by a model or a program rather than collected from human-produced sources. It is used to cover cases real data lacks, to scale instruction and preference datasets, and to avoid privacy or licensing constraints on real records. Its value depends heavily on generation quality and on filtering the results.
- [System Prompt](/en/glossary/system-prompt): A system prompt is a set of instructions supplied separately from the user's message that defines a model's role, rules, tone, and boundaries for a conversation. It is sent with every request and normally stays hidden from the end user. Models are trained to give it higher priority than user messages, though that priority is a tendency rather than a guarantee.
- [Temperature](/en/glossary/temperature): Temperature is a setting that controls how random a model's output is. At each step the model produces a probability for every possible next token, and temperature reshapes that distribution before one token is chosen. Low values concentrate probability on the most likely candidate, producing consistent output, while high values flatten it and admit less likely choices.
- [Token](/en/glossary/token): A token is the unit of text a language model actually reads and writes. It is usually a common word, a word fragment, a punctuation mark, or a space plus a word. Models process sequences of tokens rather than characters or whole words, and usage is normally metered per token, so token counts determine both cost and how much fits in one request.
- [Tokenization](/en/glossary/tokenization): Tokenization is the step that converts raw text into the sequence of tokens a model can process, and converts the model's output back into text. Modern systems use subword algorithms that learn a fixed vocabulary from training data, keeping frequent words whole and splitting rare ones into pieces. It happens before any model computation and is invisible in the final result.
- [Top-p Sampling](/en/glossary/top-p-sampling): Top-p sampling, also called nucleus sampling, limits the model's choice at each step to the smallest set of candidate tokens whose probabilities add up to a threshold p. One token is then drawn from that set alone. Unlike a fixed cutoff, the set size adapts: it stays small when the model is confident and grows when many continuations are plausible.
- [Training Data](/en/glossary/training-data): Training data is the collection of text, code, images, or other content a model learns from during training. Its composition, quality, filtering, and licensing shape what a model knows, how it writes, and which biases it carries. Data selection is now regarded as comparable in importance to architecture and scale.
- [Transfer Learning](/en/glossary/transfer-learning): Transfer learning is the practice of reusing a model trained on one task or dataset as the starting point for a different, usually narrower task. Rather than learning from random initialization, the new task inherits representations already learned from broad data. It is the general principle that makes the pre-train then adapt pipeline possible.
- [Transformer](/en/glossary/transformer): The transformer is the neural network architecture behind nearly all current language models. Introduced in a 2017 research paper, it processes a whole sequence at once and uses a mechanism called attention to let every position weigh the relevance of every other position. That design parallelizes well on modern hardware, which is what made training on internet-scale text practical.
- [Turing Test](/en/glossary/turing-test): The Turing test is a proposed criterion for machine intelligence, introduced by Alan Turing in 1950 as the imitation game, in which a human judge holds text conversations with a person and a machine and tries to tell them apart. If the judge cannot reliably do so, the machine is said to pass. Its status as a measure of intelligence is contested.
- [Unsupervised Learning](/en/glossary/unsupervised-learning): Unsupervised learning is a machine learning approach that finds structure in data carrying no labels. Instead of predicting a known answer, algorithms group similar items, reduce dimensionality, estimate density, or detect items that do not fit the observed pattern. Results describe the data rather than being correct or incorrect against a supplied target.
- [Zero-shot Prompting](/en/glossary/zero-shot-prompting): Zero-shot prompting asks a model to perform a task from instructions alone, with no worked examples included in the prompt. It relies on abilities acquired during pre-training and instruction tuning. Modern instruction-tuned models handle a wide range of common tasks this way, which keeps prompts short and cheap, though results vary more on unusual output formats.

[Back to the AI Glossary](/en/glossary)