OptDetect identifies low-optimization native libraries in Android apps with 81.9% real-world accuracy, finding the issue in 30.5% of libraries across 91.7% of top apps, with fixes yielding 10-63% CPU reduction.
Modern code reviews in open-source projects: which problems do they fix?
3 Pith papers cite this work. Polarity classification is still indexing.
fields
cs.SE 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
Eye-tracking experiment finds that labeling code as LLM-generated increases fixation time without changing review thoroughness, with reviewers adapting criteria or using the prompt.
LLM-based Go code review reaches 28% refinement exact match by switching to issue-list generation with neighboring and co-change context augmentation, outperforming primary-issue baseline and CodeReviewer.
citing papers explorer
-
Source-Free Detection and Impact Analysis of Compiler Optimization Problems in Mobile Applications
OptDetect identifies low-optimization native libraries in Android apps with 81.9% real-world accuracy, finding the issue in 30.5% of libraries across 91.7% of top apps, with fixes yielding 10-63% CPU reduction.
-
Same Scrutiny, More Time: Eye Tracking Insights into Reviewing LLM-Labelled Code
Eye-tracking experiment finds that labeling code as LLM-generated increases fixation time without changing review thoroughness, with reviewers adapting criteria or using the prompt.
-
Improving LLM-Based Go Code Review through Issue-List Generation and Context Augmentation
LLM-based Go code review reaches 28% refinement exact match by switching to issue-list generation with neighboring and co-change context augmentation, outperforming primary-issue baseline and CodeReviewer.