Demystifying Deep Learning Compiler Front End Bugs: An LLM-Aided Empirical Study
The paper presents an empirical study of frontend bugs in deep learning compilers, focusing on TorchDynamo for PyTorch 2. Using a domain‑knowledge‑enhanced LLM‑aided approach, the authors analyzed 123 bugs and built a taxonomy with seven root cause categories and fifteen subcategories. They also generated targeted test cases that uncovered 23 previously unknown bugs, confirming 15 of them, demonstrating the method’s effectiveness for testing compiler frontends.
- ▪The study examines defects introduced during the frontend translation of deep learning programs into graph‑based intermediate representations, termed fBugs.
- ▪A taxonomy of seven root cause categories and fifteen subcategories was constructed after analyzing 123 fBugs in TorchDynamo.
- ▪The authors employed a large language model to create root cause‑aware test cases, leading to the discovery of 23 new fBugs, with 15 confirmed across eight subcategories.
- ▪The findings provide actionable insights for improving development and testing practices of deep learning compiler frontends.
Hacker News (AI / LLM) files mainly under ai. We currently carry 3,301 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv.org |
| Canonical URL | https://arxiv.org/abs/2607.25651 |
| Publication time | Mon, 03 Aug 2026 03:01:29 +0000 |
| Retrieval time | 2026-08-03T03:05:43.772Z |
| Last seen | 2026-08-03T03:05:43.772Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | bl0iBysoxaKz · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Programming Languages arXiv:2607.25651 (cs) [Submitted on 28 Jul 2026] Title:Demystifying Deep Learning Compiler Frontend Bugs: An LLM-Aided Empirical Study Authors:Xinyi Yuan, Wei Chen, Jinyi Liu, Pengyu Chen, Jun Wei, Guoquan Wu, Jiaxin Zhu, Tao Huang View a PDF of the paper titled Demystifying Deep Learning Compiler Frontend Bugs: An LLM-Aided Empirical Study, by Xinyi Yuan and 7 other authors View PDF HTML (experimental) Abstract:Deep learning compilers (DLCs) are designed to translate deep learning programs into optimized, hardware-specific code. Typically, DLC frontends translate programs into graph-based intermediate representations (IRs) to enable optimizations.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
