Newest papers

LoopHIL: Looping Hardware into LLM-assisted Agents for Wireless Embedded Systems

Dhairya Shah, Spanddhana Sara, Savitha Viswanadh Kandala, Ambuj Varshney · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.24231 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Agen LLM banyak digunakan untuk pengembangan perangkat lunak dan debugging, tetapi adopsi mereka dalam sistem tertanam, terutama untuk tugas nirkabel, tetap terbatas. Ini karena pengembangan tertanam membutuhkan pengujian hardware-in-the-loop. Agen tidak dapat secara langsung mengamati spektrum, probe pins, atau mengukur daya, sehingga mereka tidak dapat memverifikasi fungsi tautan nirkabel; saat ini, insinyur manusia melakukan verifikasi ini. Kami secara sistematis mengevaluasi apakah agen LLM dapat membangun tautan nirkabel ketika disediakan dengan akses perangkat keras. Kami memperkenalkan LoopHIL, kerangka kerja yang memungkinkan agen mengontrol bench RF multi-perangkat, memonitornya dengan analis spektrum, analis logika, dan pengukur energi, dan memvalidasi tautan terhadap tangga verifikasi, semua dalam batas-batas keamanan yang

Machine translation (Meta M2M100); the English original follows.

Abstract

LLM agents are widely used for software development and debugging, but their adoption in embedded systems, especially for wireless tasks, remains limited. This is because embedded development requires hardware-in-the-loop testing. Agents cannot directly observe the spectrum, probe pins, or measure power, so they cannot verify wireless link functionality; currently, human engineers perform this verification. We systematically evaluate whether LLM agents can establish wireless links when provided with hardware access. We introduce LoopHIL, a framework that lets agents control a multi-device RF bench, monitor it with a spectrum analyzer, logic analyzer, and energy meter, and validate links against a verification ladder, all within code-enforced safety limits. Across four increasingly complex tasks, from generating a carrier to backscatter decoding by a commercial receiver, and using six models from two vendors, hardware-in-the-loop increases task completion rates from 54% to 79% and from 17% to 83% on the most challenging task. It also halves task time and output tokens. With debugging tools, agents decrease tag power consumption by up to 96%. However, small models often report unverified claims, so independent hardware checks remain necessary.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Debugging, Wireless, Computer science, Embedded system, Task (project management), Power consumption

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange