From 81ef0558fdc7b126cad9178702ca4ece3af0bddb Mon Sep 17 00:00:00 2001 From: JavaTailor Date: Wed, 8 Sep 2021 08:57:24 +0800 Subject: [PATCH] update --- README.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 8e801f2bc..d1b022c86 100755 --- a/README.md +++ b/README.md @@ -8,15 +8,14 @@ In this work, we propose JavaTailor, a history-driven test program synthesis tec The overview of JavaTailor is shown below, which consists of two main phase: the Synthesis Phase and the Execution Phase. The input of JavaTailor is historical test programs and seed programs. The output is the difference report of JVMs behaviors and the corresponding synthesized program that expose these differences. - -![01-overview](./png/01-overview.png) +01-overview

Figure 1: Overview of JavaTailor

Experimental results on popular JVM implementations (i.e., HotSpot and OpenJ9) show that JavaTailor outperforms the state-of-the-art technique in generating more diverse and effective test programs, e.g., test programs generated by JavaTailor can achieve higher JVM code coverage and detect many more unique inconsistencies than the state-of-the-art technique. For example, Figure 2 shows the coverage comparison results, where the gray lines present the coverage achieved by the seed programs, and the green and yellow lines present the coverage achieved by JavaTailor and classming on the basis of the seed programs respectively. -![05-compare](./png/05-compare.png) +05-compare

Figure 2: JVM code coverage comparison