LOCUS: Task-Aware Low-Rank Post-Training for Token-Efficient Language Generation

2026-09-10

AI Summary

本文提出LOCUS方法，通过选择任务感知的低秩适应子空间来减少语言模型生成长度，同时保持效用不变，从而降低服务成本。

Abstract

Large language model serving costs scale directly with output sequence length, yet standard preference alignment often inflates response verbosity without improving utility. We study whether the parameterization of post-training updates affects generation length: low-rank subspaces alter sequence length without modifying the alignment loss. We present LOCUS, a method that selects a task-aware low-rank adaptation subspace to minimize output-token cost subject to a utility constraint. Within this subspace, post-training retains the native preference objective with a frozen backbone. Across Anthropic HH-RLHF dialogue preferences, we evaluate two $\sim$3B decoder backbones, Pythia-2.8B and Qwen2.5-3B, against protocol-matched full-parameter DPO and DrDPO branches and the released SamPO checkpoint. LOCUS reduces continuation length by up to 39.84\% on Pythia-2.8B and by 14.87--17.58\% on Qwen2.5-3B while updating only 0.24--0.28\% of model parameters, with no material change in the internal preference diagnostic.

Problem

• large language model
• output sequence length
• preference alignment
• post-training updates
• generation length

Innovation

• Low-Rank Adaptation
• Task-Aware
• Token-Efficient
• Post-Training