Entropy-informed Decoding: Adaptive Information-Driven Branching

TL;DR

EDEN is an adaptive decoding framework for large language models that dynamically allocates computational effort based on output uncertainty, leading to improved quality and efficiency over traditional methods.

Contribution

Introduces EDEN, a model-agnostic, entropy-informed decoding method that adaptively adjusts branching based on uncertainty, outperforming fixed-width beam search.

Findings

EDEN improves output quality across mathematical reasoning, code generation, and scientific questions.

EDEN achieves better accuracy-expansion trade-offs than fixed-width beam search.

Theoretical guarantees show monotone entropy-based branching finds more probable continuations.

Abstract

Large language models (LLMs) achieve remarkable generative performance, yet their output quality is dependent on the decoding strategy. While sampling-based methods (e.g., top-k, nucleus) and search-and-select based methods (e.g., beam search, best-of-n, majority voting) can improve upon greedy decoding, both approaches suffer from limitations: sampling generally commits to a single path, while search often expends excessive computation regardless of task complexity. To address these, we introduce Entropy-informed decoding (EDEN), a plug-and-play, model-agnostic decoding framework that adaptively allocates computation based on the model's own uncertainty, approximating higher-width beam search with fewer expansions. At each generation step, EDEN estimates the entropy of the output token distribution and adjusts the branching factor monotonically with the entropy, expanding more candidates in high-entropy regions and following a greedier path in low-entropy regions, improving token efficiency. Experiments across complex tasks, including mathematical reasoning, code generation, and scientific questions, demonstrate that EDEN consistently improves output quality over existing decoding strategies, achieving better accuracy-expansion trade-offs than fixed-width beam search. By treating next-token selection as a noisy maximisation problem, we prove that branching factors monotone in entropy are guaranteed to find better (i.e. more probable) continuations than any fixed branching factor within the same total expansion budget, and derive explicit regret rates characterising the benefit of the adaptive allocation.