Don't Search for a Search Method -- Simple Heuristics Suffice for Adversarial Text Attacks

TL;DR

This paper demonstrates that simple heuristics outperform complex search algorithms in adversarial NLP attacks, questioning the effectiveness of current benchmarks and constraints.

Contribution

It shows that straightforward heuristic methods are highly effective in black-box adversarial text attacks, challenging the reliance on complex optimization techniques.

Findings

Heuristics outperform search algorithms in constrained setups.

Optimization methods offer limited benefits in constrained scenarios.

Current benchmarks are too easy and overly strict.

Abstract

Recently more attention has been given to adversarial attacks on neural networks for natural language processing (NLP). A central research topic has been the investigation of search algorithms and search constraints, accompanied by benchmark algorithms and tasks. We implement an algorithm inspired by zeroth order optimization-based attacks and compare with the benchmark results in the TextAttack framework. Surprisingly, we find that optimization-based methods do not yield any improvement in a constrained setup and slightly benefit from approximate gradient information only in unconstrained setups where search spaces are larger. In contrast, simple heuristics exploiting nearest neighbors without querying the target function yield substantial success rates in constrained setups, and nearly full success rate in unconstrained setups, at an order of magnitude fewer queries. We conclude from these results that current TextAttack benchmark tasks are too easy and constraints are too strict, preventing meaningful research on black-box adversarial text attacks.