Natural Language Processing

Computational processing of human language.

foundation tier

Natural Language Processing addresses computational processing of human language. It sits within AI and Machine Learning and inherits that area’s core questions about correctness, scale, and tractability. This page surveys the conceptual axes of the topic and points to the references that frame ongoing research and teaching. The intent is to be useful both as an entry point for newcomers and as an index for practitioners cross-checking their mental model against the field’s primary sources.

Work on natural language processing can be organised around a few interlocking concerns: the formal objects under study, the algorithms or systems that compute over them, the resource trade-offs (time, memory, communication, statistical efficiency), and the empirical or theoretical guarantees that practitioners rely on. The sources cited below approach the topic from a mix of these angles.

Foundational references

Jurafsky, Speech and Language Processing (2024) is a standard reference for this material and is used both as a curriculum anchor and as a long-form survey of techniques.

Historical context

Attention Is All You Need (Vaswani, 2017) situates the topic in its historical trajectory; revisiting it clarifies which ideas in current practice are recent and which trace back to the field’s founding texts.

Open methodological questions in natural language processing cluster around how to compose the techniques above under realistic constraints — scale, adversarial inputs, partial observability, and shifting workloads. The cited references give the precise statements, proofs, and empirical evaluations that this overview only sketches; downstream topic pages drill into specific subfields.

Prerequisites

Sources

In context

Where this topic sits in the prerequisite graph. Click any node to jump.

Open in full atlas →

Reviewed by

Explore

  1. 01

    Retrieval-Augmented Generation

    Augmenting language models with non-parametric memory — retrieving evidence at inference time from documents, knowledge graphs, or external stores to ground generation in verifiable, up-to-date information.
  2. 02

    Text Classification

    Sentiment, topic, and intent classification of text.
  3. 03

    Language Model Fairness

    The study of how large language models reproduce, amplify, or attenuate social biases — including dialect, gender, race, and intersectional bias — and the methods used to audit and mitigate them.
  4. 04

    Named Entity Recognition

    Sequence labeling for entity extraction.
  5. 05

    Language Model Alignment

    The methods used to train large language models to behave helpfully and harmlessly — reward models, RLHF, direct preference optimization, AI feedback, refusal training, and norm elicitation.
  6. 06

    Speech Language Models

    Treating speech as a sequence of discrete tokens generated by language-model-style architectures — neural codecs, speech tokenization, and TTS recast as next-token prediction over audio.
  7. 07

    Syntactic Parsing

    Constituency and dependency parsing.
  8. 08

    Multimodal Language Models

    Language models that ingest and emit non-text modalities — images, video, audio, document layout — via shared encoders, cross-attention, or projector layers, alongside the alignment, prompting, and hallucination problems unique to that joint setting.
  9. 09

    Jailbreak and Red Teaming

    Adversarial methods for breaking language model alignment — black-box and white-box jailbreaks, multimodal attacks, nested-prompt attacks, and the systematic study of safety failure modes.
  10. 10

    Semantic Parsing

    Mapping natural language to logical or executable forms.
  11. 11

    Machine Translation

    Statistical and neural translation between languages.
  12. 12

    Text Summarization

    Extractive and abstractive summarization.
  13. 13

    Dialogue Systems

    Task-oriented and open-domain dialogue agents.
  14. 14

    Word Embeddings

    Word2Vec, GloVe, and other static word vector models.
  15. 15

    Contextual Embeddings

    ELMo, BERT-style contextual representations.
  16. 16

    Language Models

    Statistical and neural language models from n-grams to transformers.
  17. 17

    Large Language Models

    Pretraining, scaling, and capabilities of large LMs.
  18. 18

    Language Model Pretraining

    Pretraining objectives, data curation, and scaling laws.
  19. 19

    Instruction Tuning

    Supervised fine-tuning on instruction-following data.
  20. 20

    Prompt Engineering

    Prompting strategies, few-shot, and chain-of-thought.
  21. 21

    Chain-of-Thought Reasoning

    Step-by-step reasoning prompts and reasoning models.
  22. 22

    Tool Use by LLMs

    Function calling, tools, and API use in language models.
  23. 23

    Parameter-Efficient Fine-Tuning

    LoRA, adapters, and prompt tuning.
  24. 24

    LLM Inference Optimization

    KV-cache, speculative decoding, and serving optimizations.
  25. 25

    Long-Context Models

    Extending context length and long-context attention.
  26. 26

    LLM Evaluation

    Benchmarks, holistic evaluation, and capability elicitation.
  27. 27

    Hallucination and Truthfulness

    Measuring and reducing fabricated outputs in LMs.
  28. 28

    Multilingual NLP

    Cross-lingual transfer and low-resource languages.
  29. 29

    Code Language Models

    LLMs trained on source code for completion and reasoning.
  30. 30

    Speech Recognition

    Acoustic and end-to-end speech-to-text systems.
  31. 31

    Speech Synthesis

    Neural text-to-speech and voice cloning.

Review this topic

This page was drafted by an agent and is waiting on expert review. Spotted a wrong prerequisite, a missing concept, a misattributed source, or a factual slip? Tell us — your review opens a tracked issue maintainers act on.