Jongwook Han

PhD Student at @SeoulNatlUni
Research interest: Value alignment in Robots & LLMs.
Lab member of :

[Jongwook Han @jwhansnu]

@taka8hiroshi Congratulations!
It must have been a hard time

[Jongwook Han @jwhansnu]

@chris_j_paxton Interesting! Thankyou for sharing

[Jongwook Han @jwhansnu]

@PeterYChong Nice work! ⚽️

[Jongwook Han @jwhansnu]

@Melletenmoon Thank you :) See you at the poster session!

[Jongwook Han @jwhansnu]

#ICML2026 If you are interested in value alignment, mechanistic interpretability, or pluralistic values, come see our poster @ JUL/8 (Wed)!

Time: 10:30
Location: Hall A, #3410
Title: Dual Mechanisms of Value Expression: Intrinsic vs. Prompted Values in Large Language Models

[Jongwook Han @jwhansnu]

@cloneofsimo Lucio available

[Jongwook Han @jwhansnu]

@AntonioSitongLi Nice work!
Would love to see Nori integrated into LeRobot.

[Robotics Papers @OWW]

RobotValues: Evaluating Household Robots When Human Values Conflict

Jongwook Han, Hyeongjin Kim, Yohan Jo
https://t.co/fmPNOb9228 [𝚌𝚜.𝚁𝙾 𝚌𝚜.𝙰𝙸]

[Jongwook Han @jwhansnu]

📄 Value Portrait: Understanding Values of LLMs with Human-aligned Benchmark

[Jongwook Han @jwhansnu]

📢 New paper @ ACL 2025 Main!

TL;DR: What humans or LLMs think a text reflects ≠ how actual value-holders respond.

🔍 We propose a psychometrically grounded way to evaluate LLM values.

📄 https://t.co/o3cCI3sCLu

#ACL2025NLP #ValueAlignment #Values #LLM #Benchmark