2 papers
cs.CL2024
Unlocking Decoding-time Controllability: Gradient-Free Multi-Objective Alignment with Contrastive Prompts
Tingchen Fu, Yupeng Hou, Julian McAuley +1
The task of multi-objective alignment aims at balancing and controlling the different alignment objectives (e.g., helpfulness, harmlessness and honesty) of large language models to…
cs.CL2024
Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction
Tingchen Fu, Deng Cai, Lemao Liu +2
Supervised fine-tuning (SFT) on instruction-following corpus is a crucial approach toward the alignment of large language models (LLMs). However, the performance of LLMs on standar…