Techniques & Methods
Greedy Algorithms
Greedy algorithms select the best available option at each decision point without considering future consequences. In text generation, greedy decoding always selects the highest-probability next token, which is fast but can miss globally optimal sequences that require temporarily choosing a lower-probability token.
Greedy decoding often produces repetitive or locally optimal but globally suboptimal text. Beam search, sampling, and nucleus sampling are alternatives that explore multiple candidates to find better overall sequences.
Authority Links
Related Terms
Techniques & Methods
Beam Search
Search algorithm that maintains multiple candidate sequences to find high-quality generated outputs.
Techniques & Methods
Decoding Rules
Guidelines and algorithms that control how language models translate internal representations into output tokens.
Techniques & Methods
Sequence Generation
Process where models produce sequences—such as words or tokens—based on learned patterns.
Techniques & Methods
Heuristics
Practical problem-solving approaches using rules of thumb rather than exhaustive search.