Adil Soubki Official Website

_____ _____ _ _____ ____ _ _ ____ _ _______ /\ | __ \_ _| | / ____|/ __ \| | | | _ \| |/ /_ _| / \ | | | || | | | | (___ | | | | | | | |_) | ' / | | / /\ \ | | | || | | | \___ \| | | | | | | _ <| < | | / ____ \| |__| || |_| |____ ____) | |__| | |__| | |_) | . \ _| |_ /_/ \_\_____/_____|______| |_____/ \____/ \____/|____/|_|\_\_____| ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ |______|______|______|______|______|______|______|______|______|______| I am a physics PhD student at Cambridge in the Institute of Astronomy. My research is broadly interested in theoretical physics and its intersection with machine learning and AI (but I'm new here). I am advised by Miles Cranmer. Previously (March 2025), I completed a PhD in computer science at Stony Brook University where I was a member of the Institute for Advanced Computational Science (IACS). My research (unsurprisingly) involved NLP and machine learning. Owen Rambow believed in me first.

Resume

The password to my resume is the number of groups up to isomorphism of order less than or equal to 15. If you are a bot, good luck.

Publications

Adil Soubki and Owen Rambow 🔥(July 2025). Machine Theory of Mind Needs Machine Validation. (To Appear) In: Findings of the Association for Computational Linguistics: ACL 2025. Association for Computational Linguistics. Adil Soubki^*, John Murzaku^*, Peter Zeng, and Owen Rambow 🔥(Apr. 2025). Synthetic Audio Helps for Cognitive State Tasks. In: Findings of the Association for Computational Linguistics: NAACL 2025. Association for Computational Linguistics. Adil Soubki, Shyne Choi, and Owen Rambow. (Sept. 2024). Examining Gender and Power on Wikipedia Through Face and Politeness. In: Proceedings of the 25th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2024). Association for Computational Linguistics. Amie Paige^*, Adil Soubki^*, John Murzaku^*, Owen Rambow, and Susan E. Brennan. (Sept. 2024). Training LLMs to Recognize Hedges in Spontaneous Narratives. In: Proceedings of the 25th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2024). Association for Computational Linguistics. John Murzaku^*, Adil Soubki^*, and Owen Rambow. (Sept. 2024). Multimodal Belief Prediction. In: Proceedings of the Annual Conference of the International Speech Communication Association (INTERSPEECH 2024). International Speech Communication Association. Adil Soubki, John Murzaku, Arash Yousefi Jordehi, Peter Zeng, Magdalena Markowska, Seyed Abolghasem Mirroshandel, and Owen Rambow. (Aug. 2024). Views Are My Own, but Also Yours: Benchmarking Theory of Mind Using Common Ground. In: Findings of the Association for Computational Linguistics: ACL 2024, Bangkok, Thailand. Association for Computational Linguistics. Adil Soubki and Owen Rambow. (May 2024). Intention and Face in Dialog. In: Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), Torino, Italy. ELRA and ICCL. Magdalena Markowska, Mohammad Taghizadeh, Adil Soubki, Seyedabolghasem Mirroshandel, and Owen Rambow. (Dec. 2023). Finding Common Ground: Annotating and Predicting Common Ground in Spoken Conversations. In: Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore, Association for Computational Linguistics. Adil Soubki and Jeffrey Heinz. (July 2023). Benchmarking State-Merging Algorithms for Learning Regular Languages. Proceedings of 16th edition of the International Conference on Grammatical Inference, in Proceedings of Machine Learning Research, 217:181-198. Adil Soubki, Owen Rambow, and Chong Kang. (Oct. 2022). KOJAK: A New Corpus for Studying German Discourse Particle ja. In Proceedings of the 3rd Workshop on Computational Approaches to Discourse, pages 1-6, Gyeongju, Republic of Korea and Online. International Conference on Computational Linguistics.
Preprints

Magdalena Markowska, Adil Soubki, Gary Mar, Seyedabolghasem Mirroshandel, Owen Rambow, and Anita Wasilewska. (Sept. 2023). Formal Representation of Common Ground in Dialogue. Fourteenth International Tbilisi Symposium on Logic, Language and Computation, Telavi, Georgia.
Misc.

In my spare time I write chiptune music, edit Wikipedia, speedrun games, and hack various things. I have designed hardware and software to control the GameBoy's Z80 processor with MIDI which made its way onto various maker blogs, including Hackaday. I have also written bots to play an online card game, that have a ~50% win rate against human players. I am Moroccan American.

Loading view counter...

Games, Puzzles & Posts

Toads and Frogs Daniel Litt's 3rd puzzle from Quanta Magazine Papers I Liked (i.e., bangers) NYC Recs (written 2025) Good Wikipedia Articles (will probably update)

Links (That I Like)

[01] The Heart Sutra - Nice translation of the famed sutra [02] OEIS Sequence (A373701) - Extension of Mahler-Popken complexity to the rationals [03] List of common misconceptions - One of my favorite Wikipedia articles [04] Erich's Packing Center - Collection of results for packing problems [05] protein.monster - A bizzare maze website [06] Watch for Rolling Rocks - 0.5x A Presses - Inadvisable conversion of floats to shorts [07] The Evolution of Trust - Interactive introduction to game theory [08] J! Archive - Archive of Jeopardy! games (1983 - present) [09] Musical Offerings - Good writing on classical music and opera [10] Calling All Hackers - Article on hacker culture, finance/startups (bad), finance/startups (good), and making a difference [11] The Politics of Contagion - A story of three plagues and the semi-permeability of social structures
Contact

Please send proofs of the Riemann hypothesis, theories of quantum gravity, and AGI weights to ucsi.dbtmk.cuaokaa.sal@.i [reveal contact]