Lightnews — Scholar-powered news

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 10d

Meet Soha, another proceedings co-chair at #ASE2025 💫
She’s an Assistant Professor at Ain Shams University, Egypt, and is passionate about all aspects of building reliable programs. ⌨️ Thrilled to have her on board for ASE'25!

1 2

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 10d

Meet Yannic, Proceedings Co-Chair of #ASE2025!
He’s helping manage the publication process together with Soha Hussein.
💪 Co-organizer of #FUZZING’26
👨‍💻 Passionate about improving software quality
@yannicnoller.bsky.social

1 7

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 10d

The #ASE25 venue, Grand Walkerhill Seoul, is located on the eastern side of Seoul, right next to the Han River. Nearby, you’ll find Achasan Mountain, a lovely spot for a short hike. If you have some free time, take a walk up with your colleagues and enjoy the beautiful panoramic view of Seoul 💙

1 1 3

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 10d

You can reach the venue by taxi or public bus, or take advantage of the free shuttle bus that runs every 15 minutes from Gangbyeon Station (Line 2) or Gwangnaru Station (Line 5).

More info: www.walkerhill.com/grandwalkerh...

GRAND

GRAND WALKERHILL SEOUL, DOUGLAS HOUSE, BOOK A ROOM, BOOK A TABLE, WALKERHILL REWARDS, WALKERHILL PRESTIGE CLUB, MEMBERSHIP, MEETING, WEDDING, ASTON-HOUSE, FAMILY GATHERINGS, BRIDE AND GROOM MEETING, ...

www.walkerhill.com

1 2

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 18d

Meet Ben, Doctoral Symposium Co-Chair & D&I Chair at #ASE2025 — his research focuses on secure software engineering and yes, he sails 😎⛵!

1 1 5

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 26d

Here’s our other Program Co-Chair, Lingming Zhang (@llingming.bsky.social) from UIUC! “Playing with software” sounds like the perfect spirit for ASE! :-)

ASE2025 Program Co-Chair

Lingming Zhang (he/him),
Prof at UIUC 🌽, USA 🇺🇸

🎮 Make a living by playing with software — synthesis, hack, and repair; neural or symbolic.

1 3

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 27d

Meet Marcel Böhme (@mboehme.bsky.social), our Program Co-Chair of ASE 2025! 😉 Together with his fellow chair, Lingming Zhang (UIUC), he is overseeing the Research Track submissions.

Marcel Böhme (he/him),
Scientist at MPI, Germany 🇩🇪

ASE'25 Program Co-Chair

🦾 #FUZZING’26 co-organizer,
🌱 Open Science enthusiast,
Father of two daughters,
🏃‍♂️ 15km every day.

1 1 5

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 24d

Applications Open: ASE 2025 Student Volunteers

🕐 ~20 hrs of service during conference week
🎟️ Free ASE registration (SV covers student reg fee; if you have a paper, one co-author must register at full rate)
👕 ASE 2025 T-shirt + more!

Please share this with your students and networks!

Apply here ⤵️

ASE 2025 - Student Volunteers - ASE 2025

Welcome to the website of the 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025. The ASE conference is the premier research forum for Automated Software Engineering. E...

conf.researchr.org

1 2

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 22d

Meet Sang Kil, one of our NIER Track Co-Chairs for #ASE2025! Seems like we’ve got quite a few runners on the ASE’25 OC team 😉

1 3

Reposted by Shin Yoo

Jonathan Aldrich @jonathanaldrich.bsky.social · 26d

Canada signed on to ACM OPEN! I'm super happy, this was worrying to a lot of my Canadian colleagues. Now faculty and students at Canadian universities that are part of the consortia can publish Open Access with ACM with no article processing charges.

www.acm.org/media-center...

ACM Signs Transformative ACM Open Agreement with Canadian Regional Consortia

ACM, the Association for Computing Machinery, the world’s largest computing society, has signed a transformative read-and-publish agreement with a group of 8 Canadian regional library consortia, repre...

www.acm.org

2 2 12

Shin Yoo @drsyoo.bsky.social · 27d

I would like to add that @mboehme.bsky.social runs 15Km pert day only because I let him have my 6Km's worth of running for free 🫢😛

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · 27d

Meet Marcel Böhme (@mboehme.bsky.social), our Program Co-Chair of ASE 2025! 😉 Together with his fellow chair, Lingming Zhang (UIUC), he is overseeing the Research Track submissions.

7

Reposted by Shin Yoo

Physicians for a Healthy Democracy @physiciandemocracy.medsky.social · Sep 4

In 1900, 31% of all deaths were children under the age of 5. Today it’s 0.7%. Vaccines, medications, and modern science are the reason why. Trust the science, it’s been proven to work safely and efficiently time and time again.

450 7K 20K

Shin Yoo @drsyoo.bsky.social · Jun 21

Off to Trondheim. Hope to see many people there!

4

Shin Yoo @drsyoo.bsky.social · Jun 10

I've been joking about how I wrote a first-author paper after such a long time. Now the reality is hitting back, and the idea of giving a 10 minute talk is actually a bit daunting, knowing that I am prone to go over time :p

Shin Yoo @drsyoo.bsky.social · Mar 14

Here's something that I am genuinely excited about. We propose a graph-like representation for executions of LLM agents called semantic flow. Intuitively speaking, it connects clusters of latent vectors which are embeddings of LLM outputs. Will appear at FSE 2025 IVR track.

arxiv.org/abs/2503.10310

Capturing Semantic Flow of ML-based Systems

ML-based systems are software systems that incorporates machine learning components such as Deep Neural Networks (DNNs) or Large Language Models (LLMs). While such systems enable advanced features suc...

arxiv.org

2

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · Jun 3

🎉 HUGE NEWS! 🎉

#ASE2025 has received a record-breaking 1,190 submissions – the most EVER for a single-cycle top SE conference! 🤯

Massive thanks to all authors who submitted. Looking forward to what promises to be an unforgettable conference with an incredible program 🔥!

ASE 2025

Welcome to the website of the 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025. The ASE conference is the premier research forum for Automated Software Engineering. E...

conf.researchr.org

7 18

Reposted by Shin Yoo

Marcel Böhme @mboehme.bsky.social · May 3

✨️ Now that #ICSE25 is over, it's time to get your papers ready for #ASE25 (30th May)!

📢 Here is what's new:
* Major Revision v2.0
* Review criteria for tech. & experience papers
* Policy on LLM-assisted Reviews
* Auto-bidding (TPMS)
* Rapid Response Reliable Reviewers

👇 For more details, read on.

1 14 26

Reposted by Shin Yoo

International Conference on Automated Software Engineering (ASE) @aseconf.bsky.social · Apr 30

📆 One month to go!

The ASE 2025 Research Papers submission deadline is approaching: 30 May.

We welcome impactful ideas and results that advance the state of the art in software engineering. 😉

#ASE2025 #SoftwareEngineering #CallForPapers

4 8

Shin Yoo @drsyoo.bsky.social · Apr 30

- lack of access to good coffee

(caveat: I've been called a coffee snob/nerd by some)

3

Reposted by Shin Yoo

Marcel Böhme @mboehme.bsky.social · Apr 26

Feature of our #ICLR'25 Spotlight paper, just presented in Singapore.
www.mpi-sp.org/81927/news_p...

How solving classic probability problems can help computer scientists find bugs

www.mpi-sp.org

3 13

Reposted by Shin Yoo

Seongmin Lee @nimgnoeseel.bsky.social · Apr 14

Can we foresee the unseen in software testing?

I’ll share how statistical magic helps us quantify residual risk and make testing results more accountable and reliable✨Thanks for the invite! 🙏

Join my tutorial @ SBST (co-located w/ ICSE)
🖥️ sbft25.github.io
📅 28.04. 14:30 GMT-4 (live on Twitch too)

1 3

Shin Yoo @drsyoo.bsky.social · Mar 14

An earlier version of this idea has been going on for some time (hence the DeepTest paper), but it really blossomed into semantic flow during @rfeldt.bsky.social's visit to last December, assisted by Somin and Naryeong, a PhD and an MSc candidate respectively at @coinse.bsky.social 👏 ❤️

1

Shin Yoo @drsyoo.bsky.social · Mar 14

But it goes beyond that... for example, one can imagine "debugging" an inference from an LLM agent by finding a specific inference step that is an outlier. We hope that semantic flow can start the discussion of what analyses are possible for these emerging software systems, i.e. LLM agents.

1 2

Shin Yoo @drsyoo.bsky.social · Mar 14

Our earlier work has shown that such representation can be used to make a reasonably accurate prediction about whether the inference result is correct or not (arxiv.org/abs/2412.08281, to be presented at DeepTest 2025 at @icseconf.bsky.social later this year).

Lachesis: Predicting LLM Inference Accuracy using Structural Properties of Reasoning Paths

Large Language Models are increasingly used to build agents to perform more complex tasks. As LLMs perform more complicated reasoning through longer interactions, self-consistency, i.e., the idea that...

arxiv.org

1 1

Shin Yoo @drsyoo.bsky.social · Mar 14

We propose that we embed each LLM action or response, and connect these latent vectors in the order of their appearance during executions. When done for multiple executions, you can see shapes of the inference/reasoning chains.

1 1

Shin Yoo @drsyoo.bsky.social · Mar 14

LLMs compute along the data flow, whereas traditional code implements its semantics along the control flow. Seen from the traditional perspective, the execution of LLM agents is just a big loop of sending and receiving REST API calls (okay, I've simplified it significantly, but you get the drift).

1 1