Sheikh Shafayat

About Me

Hi, I am an M.S. student at the KAIST School of Computing, advised by Alice Oh.

My current research interests include LLM reasoning (RLVR), AI creativity, and open-endedness. I closely collaborate with Andrea Zanette for projects related to LLM reasoning and self-improvement. I have finished my undergraduate at KAIST when I worked on evaluation, multilinguality, and cultural NLP.

A (partially) maintained CV of mine can be found here.

I am really open to meeting and chatting with new people. Feel free to send me an email if you want to chat about something (any paper, potential collaboration, a cool novel, or just about anything). If you are in Seoul and Daejeon, shoot me an email to grab a coffee or go climbing together!

On my free time, I enjoy bouldering, running, or basking in the sunset with a novel.

Research Interests

NLP: LLM reasoning, algorithmic creativity, open endedness, multilinguality

News

[May. 2025] New Preprint: Can Large Reasoning Models Self-Train?
[Dec. 2024] New Preprint: A 2-step Framework for Automated Literary Translation Evaluation: Its Promises and Pitfalls
[July. 2024] MultiFAct for Multilingual Factuality got accepted to COLM-2024
[May. 2024] BEnQA got accepted to ACL Findings
[Apr. 2024] MultiFAct got accepted to ICLR Global AI Cultures Workshop

Publications

Arxiv

Can Large Reasoning Models Self-Train?

Sheikh Shafayat*, Fahim Tajwar*, Ruslan Salakhutdinov, Jeff Schneider, Andrea Zanette

PDF Code Project Page Preprint

Arxiv

A 2-step Framework for Automated Literary Translation Evaluation: Its Promises and Pitfalls

Sheikh Shafayat, Dongkeun Yoon, Woori Jang, Jiwoo Choi, Alice Oh, Seohyon Jung

PDF Preprint

COLM

Multi-FAct: Assessing Factuality of Multilingual LLMs using FActScore

Sheikh Shafayat, Eunsu Kim, Juhyun Oh, Alice Oh

Accepted to COLM 2024

PDF

Arxiv

The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Models

Seungone Kim, Juyoung Suk, Ji Yong Cho, Shayne Longpre, Chaeeun Kim, Dongkeun Yoon, Guijin Son, Yejin Cho, Sheikh Shafayat, Jinheon Baek, Sue Hyun Park, Hyeonbin Hwang, Jinkyung Jo, Hyowon Cho, Haebin Shin, Seongyun Lee, Hanseok Oh, Noah Lee, Namgyu Ho, Se June Joo, Miyoung Ko, Yoonjoo Lee, Hyungjoo Chae, Jamin Shin, Joel Jang, Seonghyeon Ye, Bill Yuchen Lin, Sean Welleck, Graham Neubig, Moontae Lee, Kyungjae Lee, Minjoon Seo

PDF Code Accepted to NAACL 2025 (Best Paper Award)

ACL

BEnQA: A Question Answering and Reasoning Benchmark for Bengali and English

Sheikh Shafayat, HM Hasan, Minhajur Rahman Chowdhury Mahim, Rifki Afina Putri, James Thorne, Alice Oh

Accepted to ACL-Findings 2024

PDF Code

ACL

LangBridge: Multilingual Reasoning Without Multilingual Supervision

Dongkeun Yoon, Joel Jang, Sungdong Kim, Seungone Kim, Sheikh Shafayat, Minjoon Seo

Accepted to ACL-Main 2024

PDF Code

Current Readings

Here are a few latest books that I am reading/just finished. I would be very grateful to receive book recommendations from anyone!

East of Eden (John Steinbeck): This is the most American thing I have ever read. A fantastic family saga relating war, violence, love, immigration, assimilation, capitalism, corruption and moral ambivalence.
Nexus (Yuval Noah Harari): Very interesting read on the role of information technology in sustaining large scale human networks. My position on this book is somewhat mixed. While I find Harari’s thesis on how information networks and intersubjective reality shape human history to be absolutely brilliant, he seems to be deeply mistaken on the current state of AI progress. I love the part I of the book, but part II? Not so much.
Never Let Me Go (Kazuo Ishiguro): Soothing, melancholic, yet dystopian. A touching exploration of love, memory, loss and meaning in life.
Meditations (Marcus Aurelius): A Stoic classic. I was always drawn to Stoicism, but never had the chance to get started with the primary sources. This book is changing that.