Skip to main content
SAICA-KG

Papers

When the Code Autopilot Breaks: Why LLMs Falter in Embedded Machine Learning

Roberto Morabito, Guanghan Wu · 2025 · arXiv

Source →arXiv:2509.10946DOI: 10.48550/arXiv.2509.109460 citations
failure-mode-taxonomycoding-agentembedded-mlempirical

TL;DR

Empirical study of LLM code-generation failure modes in embedded ML pipelines; derives a failure-category taxonomy emphasizing format-induced misinterpretations and runtime-disruptive-but-compiling code.

Abstract

Large Language Models (LLMs) are increasingly used to automate software generation in embedded machine learning workflows, yet their outputs often fail silently or behave unpredictably. This article presents an empirical investigation of failure modes in LLM-powered ML pipelines, based on an autopilot framework that orchestrates data preprocessing, model conversion, and on-device inference code generation. We show how prompt format, model behavior, and structural assumptions influence both success rates and failure characteristics, often in ways that standard validation pipelines fail to detect. Our analysis reveals a diverse set of error-prone behaviors, including format-induced misinterpretations and runtime-disruptive code that compiles but breaks downstream. We derive a taxonomy of failure categories and analyze errors across multiple LLMs, highlighting common root causes and systemic fragilities.

Editorial notes

Domain-specific failure taxonomy (embedded ML) that exposes compiles-but-breaks-downstream patterns relevant to SAICA-KG's logic_error and obsolescence modes. Narrower than the general-agent taxonomies but useful for triangulating silent-failure signals.

View raw YAML on GitHub →