Chain-of-Code Collapse: Reasoning Failures in LLMs via Adversarial Prompting in Code Generation
This exposes the fragility of reasoning in LLMs, which is a problem for AI safety and reliability, though it is incremental in investigating robustness.
The paper tackles the problem of whether LLMs truly reason or rely on shallow patterns by introducing adversarial prompt perturbations in code generation, finding that some modifications degrade accuracy by up to -42.1% while others improve it by up to 35.3%.
Large Language Models (LLMs) have achieved remarkable success in tasks requiring complex reasoning, such as code generation, mathematical problem solving, and algorithmic synthesis -- especially when aided by reasoning tokens and Chain-of-Thought prompting. Yet, a core question remains: do these models truly reason, or do they merely exploit shallow statistical patterns? In this paper, we introduce Chain-of-Code Collapse, where we systematically investigate the robustness of reasoning LLMs by introducing a suite of semantically faithful yet adversarially structured prompt perturbations. Our evaluation -- spanning 700 perturbed code generations derived from LeetCode-style problems -- applies transformations such as storytelling reframing, irrelevant constraint injection, example reordering, and numeric perturbation. We observe that while certain modifications severely degrade performance (with accuracy drops up to -42.1%), others surprisingly improve model accuracy by up to 35.3%, suggesting sensitivity not only to semantics but also to surface-level prompt dynamics. These findings expose the fragility and unpredictability of current reasoning systems, underscoring the need for more principles approaches to reasoning alignments and prompting robustness. We release our perturbation datasets and evaluation framework to promote further research in trustworthy and resilient LLM reasoning.