My colleague called LLMs βthe modern horoscopeβ, and itβs now my favorite definition π
12.02.2026 13:00 β π 1 π 0 π¬ 0 π 0@isychev.bsky.social
postdoc @TU Wien, PhD from TU Munich. Fond of program synthesis and analysis, cleaning up finite-precision mess (old spelling Anastasiia Izycheva) https://aisychev.github.io
My colleague called LLMs βthe modern horoscopeβ, and itβs now my favorite definition π
12.02.2026 13:00 β π 1 π 0 π¬ 0 π 0Congratulations, Dr.Cutler! π
09.02.2026 19:38 β π 1 π 0 π¬ 0 π 0β³ FLoC 2026 Deadlines Coming!
π Jan 15+22 (LICS)
π Jan 23+30 (FSCD)
π Jan 24+31 (ICLP)
π Jan 28 (CAV)
π Jan 29 (CSF)
π Feb 6+13 (IJCAR)
π Feb 8+13 (KR)
π Feb 12+19 (ITP)
π Feb 20+27 (SAT)
π Feb 28+Mar 7 (CP)
Learn more: www.floc26.org
#FLoC2026 #LogicInCS
My international soul with Slavic roots is delighted every time someone says I have a *German* accent when I speak English.
12.12.2025 11:42 β π 1 π 0 π¬ 0 π 0I was looking for someoneβs up-to-date email address, which led me to Linkedin. So now I down the rabbit hole updating my account that I havenβt used since 2013 π
09.12.2025 15:59 β π 3 π 0 π¬ 1 π 0say no more: open.spotify.com/track/4wgE9X... π
04.12.2025 20:31 β π 0 π 0 π¬ 0 π 0Listening to the song called βGrammarβ by the artist βFloating Pointsβ
when your professional choices spill into music choices
04.12.2025 09:56 β π 9 π 0 π¬ 1 π 0haha yeah, thatβs the spirit π
25.11.2025 19:27 β π 0 π 0 π¬ 0 π 0FSE
25.11.2025 08:18 β π 0 π 0 π¬ 0 π 0noted :)
23.11.2025 11:48 β π 0 π 0 π¬ 1 π 0Academia is such a weird thing. I already agreed to give a talk at a workshop in November 2026, but cannot tell which country Iβll be living in at that time :)
23.11.2025 11:35 β π 6 π 0 π¬ 1 π 0uh-oh, I bet you got an earful from him :)
22.11.2025 22:07 β π 2 π 0 π¬ 1 π 0I donβt know who thought itβs a good idea to start authors response period on a Friday (probably afternoon), but here we are
21.11.2025 09:45 β π 2 π 0 π¬ 2 π 0π’ Call for Submissions β FLoC 2026
Join FLoC 2026: top conferences, 86 workshops (including Mentoring Workshop, Olympic Games, tutorials, and doctoral consortia), and leading invited speakers.
π Learn more and find submission details: www.floc26.org
#FLoC2026 #LogicInCS
Two countries and almost three weeks later I finally returned home from OOPSLA :)
06.11.2025 21:20 β π 3 π 0 π¬ 0 π 0Presented our #CCS25 paper yesterday in Taipei! π€
We introduce Circuzz, the first systematic fuzzing framework for zero-knowledge (ZK) pipelines.
π Paper: mariachris.github.io/Pubs/CCS-202...
π» Repo: github.com/Rigorous-Sof...
with @isychev.bsky.social, @vwuestholz.bsky.social, Maria Christakis
On my way to #SPLASH25! Taking the longest direct flight for me so far: 12 hours. See you on the other side :)
14.10.2025 09:08 β π 2 π 0 π¬ 0 π 0Mine is October 16, 16:15, Neural Networks session :) Iβll talk about performance optimization for floats and how against all odds sound optimizers are doing great in direct comparison with unsound dynamic ones.
07.10.2025 13:30 β π 3 π 0 π¬ 1 π 0I think itβs called sabbatical :)
30.09.2025 11:26 β π 2 π 0 π¬ 0 π 0@confcav.bsky.social FOMO starting in three, two, one β¦
23.07.2025 05:09 β π 1 π 0 π¬ 0 π 0π Welcome to the official Bluesky presence of the Federated Logic Conference 2026 (FLoC 2026), coming to Lisbon, Portugal in July 2026! floc26.org #FLoC2026 #LogicInCS
18.07.2025 12:06 β π 3 π 2 π¬ 1 π 0a shot in the dark, but is anyone I know here going to IJCAI in Montreal?
01.07.2025 12:18 β π 1 π 1 π¬ 1 π 0Love the title of a studentβs report for my seminar: βFloats and Other Lies Computers Tellβ. π
20.06.2025 11:34 β π 1 π 0 π¬ 0 π 0printf gone wrong :)
31.03.2025 15:34 β π 3 π 0 π¬ 0 π 0I meant that I want to be able to finish reviews quickly, because I have a lot of other exciting things on my todo list (the institution doesnβt pressure me to do any of those)
06.03.2025 16:10 β π 0 π 0 π¬ 0 π 0Dear fellow reviewers, any tips on how to be more efficient with reviews without sacrificing quality?
I want to give detialed feedback, so spending quite a lot of time reading. Time that I sometimes don't have :)
tis the season of βreject because you donβt cite a paper that is not yet published and preprint is not availableβ (:
19.12.2024 16:34 β π 6 π 0 π¬ 1 π 0I see a big βLLLβ here, happy that itβs not βLLMβ π
02.12.2024 15:46 β π 1 π 0 π¬ 1 π 0CAV (International Conference on Computer Aided Verification) is now also on Bluesky!
Please follow @confcav.bsky.social and help us spread the word.
Videos from ICFP'24 and colocated events are available now on the ACM SIGPLAN youtube channel!
www.youtube.com/@acmsigplan/...