Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Supplement Generation Training for Enhancing Agentic Task Performance

About

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.

Young Min Cho, Daniele Bonadiman, Divya Bhargavi, Tamer Alkhouli, Salvatore Romeo, Dongwei Jiang, Khushbu Pahwa, Yubin Ge, Etsuko Ishii, Monica Sunkara, Yi Zhang• 2026

Related benchmarks

TaskDatasetResultRank
Multi-hop Question AnsweringHotpotQA--
294
ReasoningHLE
Accuracy (HLE Reasoning)5.3
63
Code GenerationDS-1000
Accuracy62
35
Text-to-SQLSpider
Accuracy82.5
28
ReasoningSuperGPQA
Accuracy (superGPQA)38.5
24
Showing 5 of 5 rows

Other info

Follow for update