Rahul GopinathAndreas Zeller

Semantic Debugging! We present AVICENNA, a tool to automatically determine precise causes + conditions of software failures: “The program fails if the given <length> is larger than the <payload> length.” Meet us today 5pm at ESEC/FSE! dl.acm.org/doi/10.1145/3611643

Rahul GopinathAndreas Zeller

In one year of work, Irina Nicolae, Max Eisele, and I have deconstructed the controversial NEUZZ ML-based fuzzer and found that its original performance claims do not hold (see also twitter.com/domenuk/status/151)
Looking forward to lively discussions today 12:15 at ESEC/FSE! dl.acm.org/doi/10.1145/3611643

Rahul Gopinath

In Zoom, I wish there was a way to watch the previous 5 or 10 minutes of the current meeting in faster speed so that I could get the context quickly or to just quickly check what was said.

Rahul Gopinath

Does anyone know how to find which apps in Mac was responsible for the last chime? Being the end of the year, and having to communicate to many teams, I have too many apps open at this time, and all the chimes sound exactly alike. #MacOS

Rahul Gopinath

At SAPLING'23 #SAPLING #SAPLING23. Thanks to Julian for capturing this image.

Rahul Gopinath

SAPLING Francois Gauthier from Oracle Labs is talking about /Runtime and Software Supply Chain Security Research at Oracle Labs/ called RASPunzel. This is of a huge interest to me as my student worked on a related project called Macaron from Oracle Labs. #SAPLING #SAPLING23

Rahul Gopinath

SAPLING also has Software Testing! Yanqi Su (ANU) is now talking about /Constructing a System Knowledge Graph of User Tasks and Failures from Bug Reports to Support Soap Opera Testing/ #SAPLING23 #SAPLING

Rahul Gopinath

For those who are attending SAPLING I am giving a talk today called "Dancing to Unknown Music: Learning Black-Box Program Input Grammars with Prefix Queries". See you there.

 

Rahul Gopinath

Running LLMs from a single file! With LLamaFile from Mozilla. All you need is 

curl -LO https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/main/llamafile-server-0.1-llava-v1.5-7b-q4
chmod 755 llamafile-server-0.1-llava-v1.5-7b-q4
./llamafile-server-0.1-llava-v1.5-7b-q4


Now, I am waiting for an easy way to finetune an LLM from a Mac (this is mostly doable now) or using a CPU, and we will soon be within reach of the world of custom agents promised by @cstross

Rahul Gopinath

Only 4 more days left to go for [SBFT'24 fuzzing competition](https://sbft24.github.io/tools/fuzzing). Please apply if you have a fuzzer that can beat the world competition!