AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent
Feb 4, 2026·
,,,,,,,·
1 min read
Yinyi Luo
Yiqiao Jin
Weichen Yu
Mengqi Zhang
Srijan Kumar
Xiaoxiao Li
Weijie Xu
Xin Chen
Jindong Wang
Abstract
Multi-agent systems achieve strong performance on complex tasks by orchestrating diverse roles, planners, and tool-using agents. However, deploying full multi-agent stacks is expensive and brittle. We introduce AgentArk, a distillation framework that compresses multi-agent intelligence into a single LLM agent. AgentArk decomposes multi-agent trajectories into role-conditioned skills and trains a single agent to reproduce the collaborative behavior of the original ensemble, recovering most of the performance at a fraction of the cost.
Type
Publication
Under Review at NeurIPS 2026 (Preprint)
Abstract
Multi-agent systems achieve strong performance on complex tasks by orchestrating diverse roles, planners, and tool-using agents. However, deploying full multi-agent stacks is expensive and brittle. We introduce AgentArk, a distillation framework that compresses multi-agent intelligence into a single LLM agent. AgentArk decomposes multi-agent trajectories into role-conditioned skills and trains a single agent to reproduce the collaborative behavior of the original ensemble.
