NCO: A Versatile Plug-in for Handling Negative Constraints in Decoding

TL;DR

NCO is a decoding strategy that efficiently enforces multiple hard and regex constraints during language model generation, reducing computational costs and supporting various inference methods.

Contribution

It introduces NCO, a novel online pattern matching decoding method that handles multiple constraints without state explosion, compatible with standard inference strategies.

Findings

Effective in suppressing PII and profanity in LLM outputs.

Reduces computational overhead compared to traditional constrained decoding.

Supports soft masking for probabilistic content suppression.

Abstract

Controlling Large Language Models (LLMs) to prevent the generation of undesirable content, such as profanity and personally identifiable information (PII), has become increasingly critical. While earlier approaches relied on post-processing or resampling, recent research has shifted towards constrained decoding methods that control outputs during generation to mitigate high computational costs and quality degradation. However, preventing multiple forbidden hard constraints or regex constraints from appearing anywhere in the output is computationally challenging. A straightforward solution is to convert these constraints into a single automaton that tracks all forbidden patterns during decoding, but this often becomes impractically large. Standard regex engines also do not readily support the operations needed to build such a constraint, such as complement and intersection. In order to address these limitations, we propose NCO, a decoding strategy that performs online pattern matching over finite hard constraints and regex constraints, reducing computational overhead without inducing state explosion. NCO is fully compatible with standard inference strategies, including various sampling methods and beam search, while also supporting soft masking for probabilistic suppression. We empirically demonstrate its effectiveness across practical tasks, including PII and profanity suppression. Our implementation is available at https://github.com/hyundong98/NCO-Decoding.git .