LoRA on the Go: Instance-level Dynamic LoRA Selection and Merging
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917421633241088 |
|---|---|
| author | Lee, Seungeon Das, Soumi Gupta, Manish Gummadi, Krishna P. |
| author_facet | Lee, Seungeon Das, Soumi Gupta, Manish Gummadi, Krishna P. |
| contents | Low-Rank Adaptation (LoRA) has emerged as a parameter-efficient approach for fine-tuning large language models. However, conventional LoRA adapters are typically trained for a single task, limiting their applicability in real-world settings where inputs may span diverse and unpredictable domains. At inference time, existing approaches combine multiple LoRAs for improving performance on diverse tasks, while usually requiring labeled data or additional task-specific training, which is expensive at scale. In this work, we introduce LoRA on the Go (LoGo), a training-free framework that dynamically selects and merges adapters at the instance level without any additional requirements. LoGo leverages signals extracted from a single forward pass through LoRA adapters, to identify the most relevant adapters and determine their contributions on-the-fly. Across 5 NLP benchmarks, 27 datasets, and 3 model families, LoGo outperforms training-based baselines on some tasks upto a margin of 3.6% while remaining competitive on other tasks and maintaining inference throughput, highlighting its effectiveness and practicality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07129 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LoRA on the Go: Instance-level Dynamic LoRA Selection and Merging Lee, Seungeon Das, Soumi Gupta, Manish Gummadi, Krishna P. Computation and Language Artificial Intelligence Machine Learning Low-Rank Adaptation (LoRA) has emerged as a parameter-efficient approach for fine-tuning large language models. However, conventional LoRA adapters are typically trained for a single task, limiting their applicability in real-world settings where inputs may span diverse and unpredictable domains. At inference time, existing approaches combine multiple LoRAs for improving performance on diverse tasks, while usually requiring labeled data or additional task-specific training, which is expensive at scale. In this work, we introduce LoRA on the Go (LoGo), a training-free framework that dynamically selects and merges adapters at the instance level without any additional requirements. LoGo leverages signals extracted from a single forward pass through LoRA adapters, to identify the most relevant adapters and determine their contributions on-the-fly. Across 5 NLP benchmarks, 27 datasets, and 3 model families, LoGo outperforms training-based baselines on some tasks upto a margin of 3.6% while remaining competitive on other tasks and maintaining inference throughput, highlighting its effectiveness and practicality. |
| title | LoRA on the Go: Instance-level Dynamic LoRA Selection and Merging |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2511.07129 |