arXiv · 2611.20635
Contrastive Activation-Guided Zeroth-Order Preference Optimisation for Memory-Efficient LLM Alignment
Abstract
Direct Preference Optimisation (DPO) simplifies reinforcement learning from human feedback, but still relies on traditional backpropagation and its associated activation-memory cost. Zeroth-order (ZO) optimisation avoids backpropagation, yet isotropic perturbations can produce noisy updates for preference objectives defined by differences between chosen and rejected responses. We introduce Contrastive Activation-guided Zeroth-Order Direct Preference Optimisation (CA-ZODPO), which constructs a low-rank search subspace from the activation discrepancy between the chosen and rejected responses in each preference pair. Perturbations are restricted to this subspace, while momentum is maintained in low-rank coordinates to preserve the memory advantages of ZO training. On Qwen2.5-1.5B with Anthropic HH-RLHF, CA-ZODPO outperforms common ZO baselines and remains competitive with first-order DPO in pairwise preference evaluation. It also reduces peak GPU memory by about 59\% at a sequence length of 2048 while retaining the sequence-length scaling behaviour of zeroth-order training.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Weiyun Huang, Alessio Galatolo. 2026-11-16. Contrastive Activation-Guided Zeroth-Order Preference Optimisation for Memory-Efficient LLM Alignment. https://searcharxiv.com/abs/2611.20635
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.