Beam search

Appears in 2 papers

An inference-time decoding algorithm.

As used in Paper 06 — Sequence to Sequence Learning with Neural Networks →

An inference-time decoding algorithm. Instead of greedily picking just

As used in Paper 23 — Scaling LLM Test-Time Compute Optimally Can be More Effective than Scaling Model Parameters →

A search algorithm that maintains the k most promising partial solutions at each step, pruning less promising ones. Balances exploration and computational cost.

Paper 06 — Sequence to Sequence Learning with Neural Networks → Paper 23 — Scaling LLM Test-Time Compute Optimally Can be More Effective than Scaling Model Parameters →

Appears in papers