cs.LGNov 8, 2026

JoMoHa: Joint Model--Harness Optimization with Reinforcement Learning

Authors: Xudong Wu, Guangchen Lan, Mingqi Yuan, Jiayu Chen

Organizations: The University of Hong Kong, Hong Kong · Salesforce AI Research

Abstract

The performance of language agents depends jointly on model weights and the harness. Optimizing either alone can be constrained by the other component, limiting overall performance gains. Our analysis of naive alternating training reveals progressive incompatibility between the model and harness, which increasingly interferes with task execution and may also undermine model training, leading to performance degradation. To coordinate the evolution of models and harnesses, we introduce JoMoHa, to our knowledge the first framework to jointly train an executable harness generation policy and a task execution policy through reinforcement learning. JoMoHa outperforms the evaluated weight-only and harness-only baselines and achieves the highest success rates among the evaluated methods on ALFWorld, WebShop, and DBBench, averaging 70.3% and reaching 97.0% on ALFWorld. By learning both task execution and runtime improvement from its own generated trajectories, JoMoHa performs recursive self-improvement (RSI).

Figures & tables

Explore similar work

CardsList
  1. WHALE: A Simple Recipe for Joint Harness-Weight Optimization

    Aug 31, 2026Haechan Kim, Yoonho Lee, Gisang Lee +2Agent Harness OptimizationLLM Agent Training

  2. Harness Annealing: Learning to Act with Less External Control

    Oct 1, 2026Yingxuan Yang, Huacan Chai, Ying WenLanguage Model-Based ControlAgent Harness Optimization

  3. HarnessBandit: Joint Learnability-Transferability Scheduling for Multi-Harness Agentic Reinforcement Learning

    Sep 12, 2026Hongliang Wei, Xiaobing Tu, Yinggui Wang +6RL for LLM AgentsAgent Harness Optimization