Supplement Generation Training for Enhancing Agentic Task Performance
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908987160526848 |
|---|---|
| author | Cho, Young Min Bonadiman, Daniele Bhargavi, Divya Alkhouli, Tamer Romeo, Salvatore Jiang, Dongwei Pahwa, Khushbu Ge, Yubin Ishii, Etsuko Sunkara, Monica Zhang, Yi |
| author_facet | Cho, Young Min Bonadiman, Daniele Bhargavi, Divya Alkhouli, Tamer Romeo, Salvatore Jiang, Dongwei Pahwa, Khushbu Ge, Yubin Ishii, Etsuko Sunkara, Monica Zhang, Yi |
| contents | Training large foundation models for agentic tasks is increasingly impractical due to the high computational costs, long iteration cycles, and rapid obsolescence as new models are continuously released. Instead of post-training massive models for every new task or domain, we propose Supplement Generation Training (SGT), a more efficient and sustainable strategy. SGT trains a smaller LLM to generate useful supplemental text that, when appended to the original input, helps the larger LLM solve the task more effectively. These lightweight models can dynamically adapt supplements to task requirements, improving performance without modifying the underlying large models. This approach decouples task-specific optimization from large foundation models and enables more flexible, cost-effective deployment of LLM-powered agents in real-world applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20727 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Supplement Generation Training for Enhancing Agentic Task Performance Cho, Young Min Bonadiman, Daniele Bhargavi, Divya Alkhouli, Tamer Romeo, Salvatore Jiang, Dongwei Pahwa, Khushbu Ge, Yubin Ishii, Etsuko Sunkara, Monica Zhang, Yi Machine Learning Artificial Intelligence Training large foundation models for agentic tasks is increasingly impractical due to the high computational costs, long iteration cycles, and rapid obsolescence as new models are continuously released. Instead of post-training massive models for every new task or domain, we propose Supplement Generation Training (SGT), a more efficient and sustainable strategy. SGT trains a smaller LLM to generate useful supplemental text that, when appended to the original input, helps the larger LLM solve the task more effectively. These lightweight models can dynamically adapt supplements to task requirements, improving performance without modifying the underlying large models. This approach decouples task-specific optimization from large foundation models and enables more flexible, cost-effective deployment of LLM-powered agents in real-world applications. |
| title | Supplement Generation Training for Enhancing Agentic Task Performance |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2604.20727 |