Skip to content
daily-hour-news·

🔬Study Finds Repair Hallucinations in 72.7% of LLM Patches

TL;DR

Manual review of 812 LLM-generated patches found repair hallucinations in 72.7% of cases, including patches that passed every available test. Only 21.0% to 55.9% of generated patches cleared the developer-written suite across three models.

Manual review of 812 LLM-generated patches found repair hallucinations in 72.7% of cases, including patches that passed every available test. Only 21.0% to 55.9% of generated patches cleared the developer-written suite across three models.

Study Finds Repair Hallucinations in 72.7% of LLM Patches — daily-hour-news

Key Points

1

arXiv 2609.04909, submitted Sep 4 by Xuemeng Cai and four co-authors

2

Three representative LLMs evaluated on 832 Defects4J bugs

3

21.0%-55.9% of patches pass the developer-written test suite

4

Incorrect causal localization drives 45.9% of hallucinations; wrong repair strategy 18.5%

5

Models also misidentify triggering test cases and mispredict line coverage on branches

Why It Matters

A green test run is not evidence your agent understood the bug, so gate auto-fixes on causal localization rather than on CI going green.

Quick Facts

program repairhallucinationDefects4JLLM evaluationsoftware engineeringarXiv

Frequently Asked Questions

Why does this matter?

A green test run is not evidence your agent understood the bug, so gate auto-fixes on causal localization rather than on CI going green.

What happened?

Manual review of 812 LLM-generated patches found repair hallucinations in 72.7% of cases, including patches that passed every available test. Only 21.0% to 55.9% of generated patches cleared the developer-written suite across three models.

Comments

Subscribe to join the conversation...

Be the first to comment

Enjoyed this article?

Get it daily. 7am. Free. Reads in 5 minutes.

Join 3,463 builders reading daily.

Also get