作者位次 #2/6ACL 2026Poster· Natural Language Processing
Self-Guided Alignment: Adaptive Preference S ensing for Multi-Objective Generation
Ning Wang · Zhanyang Liu · Taotao Zhou · Xinrui Zhang · Zongru Shao · Haojie Zhou
摘要
Aligning Large Language Models (LLMs) with diverse and potentially conflicting human values necessitates navigating complex multi-objective landscapes. However, existing prompt-conditioned approaches face a critical training-inference discrepancy: they rely on ground-truth scores during training...