The Parts Are Greater Than the Sum: Automated Task Sequencing for Efficient Training of Multi-Policy LLMs

Practical Tips

Practical tips and how-to guidance will be added by our editorial team.

中文解读

中文解读待补充：本站将优先为睡眠改善、失眠治疗、助眠方法等高价值文章补充中文说明。

Article Summary

Parameter-Efficient Fine-Tuning (PEFT) commonly adapts large language models using a single shared Low-Rank Adapter (LoRA). This shared optimization space often suffers from interference when adapting heterogeneous task sequences, leading to poor transfer and catastrophic forgetting. Existing approaches mainly improve adapter expressiveness by increasing parameter capacity or composing multiple adapters, yet they still rely on a shared optimization path. In this paper, we propose an optimization-path organization framework for parameter-efficient fine-tuning of large language models, implemented as an automatic multi-policy PEFT architecture. Specifically, optimization-compatible adaptation paths are automatically organized through task grouping and task sequencing under a fixed parameter budget. The organized optimization paths are implemented as independent Quantized Low-Rank Adapters (QLoRA), enabling heterogeneous tasks to be optimized in decoupled adaptation spaces while preserving positive transfer among compatible tasks. Experiments on the TRACE benchmark demonstrate that performance consistently improves from conventional single-policy PEFT to multi-policy PEFT, with the proposed automatic multi-policy framework achieving the best performance of 44.78 under the same trainable capacity. This suggests that optimization-path organization is more effective than simply increasing adapter capacity for heterogeneous parameter-efficient fine-tuning.

Sources & References

• Original source
• Full article (PDF)

Comments