6 papers
Meta-Learning Preferences for Multilingual LLM Alignment
Jiaying Lin, Seongho Son, Nam Phuong Tran +3
Unequal availability of human preference data across languages poses a significant challenge for aligning large language models in multilingual settings. To address the lack of suf…
SWE-Router: Routing in Multi-turn Agentic Software Engineering Tasks
Seongho Son, Sangwoong Yoon, Jiahua Tang +3
Large language models (LLMs) embedded in multi-turn agentic harnesses are reshaping software engineering (SWE), but routing every task to a frontier model is wasteful when many iss…
Overton Pluralistic Reinforcement Learning for Large Language Models
Yu Fu, Seongho Son, Ilija Bogunovic
Existing alignment paradigms remain limited in capturing the pluralistic nature of human values. Overton Pluralism (OP) addresses this gap by generating responses with diverse pers…
Robust Multi-Objective Controlled Decoding of Large Language Models
Seongho Son, William Bankes, Sangwoong Yoon +3
We introduce Robust Multi-Objective Decoding (RMOD), a novel inference-time algorithm that robustly aligns Large Language Models (LLMs) to multiple human objectives (e.g., instruct…
RSPO: Regularized Self-Play Alignment of Large Language Models
Xiaohang Tang, Sangwoong Yoon, Seongho Son +3
Self-play-based policy optimization has emerged as an effective approach for fine-tuning large language models (LLMs), formulating preference optimization as a two-player game. How…
Right Now, Wrong Then: Non-Stationary Direct Preference Optimization under Preference Drift
Seongho Son, William Bankes, Sayak Ray Chowdhury +2
Current Large Language Model (LLM) preference optimization algorithms do not account for temporal preference drift, which can lead to severe misalignment. To address this limitatio…