agent harness 1evaluation datasets 1large language models 1prompt engineering 1reinforcement learning 1tool use 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.AI2026
A Control System, a Dataset, and a Recipe for Making Frozen LLM Agents Learn a Domain
Debjyoti Paul
The paper introduces a lightweight control system that learns to optimize the prompt and tool harness of frozen large language model agents using sample‑efficient reinforcement lea…
cs.AI2026
Context Assembly as the Controlled Variable: A Control-Theoretic View of Harness Policies for Frozen LLM Agents
Debjyoti Paul
A growing body of 2026 work applies control theory to LLM agents: Lyapunov-certified stability for tool-mediated controllers (Prinos et al., "Stable Agentic Control", 2026), sample…