AI & ML Tech Glossary

Contrastive decoding improves LLM generation by combining a strong model’s likelihood with a penalty from a weaker model, discouraging generic continuations and reducing repetition without...

Continuous batching dynamically batches LLM requests as they arrive, especially during token-by-token decoding, improving GPU utilization and throughput for streaming chat and agent workloads with...

Constrained decoding restricts an LLM’s token generation to outputs that satisfy a formal constraint such as valid JSON, a schema, a grammar, or a fixed...

Chain-of-thought prompting is a technique that encourages an LLM to write intermediate reasoning steps before the final answer, which can improve multi-step problem solving and...

Contextual compression shortens retrieved documents for RAG by extracting or summarizing only query-relevant information, reducing tokens and noise while preserving evidence for grounded LLM answers.

Constitutional AI is an alignment method that uses a written set of principles to guide an LLM’s self-critique and revision during training, helping it follow...

A context window is the token limit a language model can attend to in a single request. It determines how much conversation and document text...