Harness Learning Enables Generalizable Test-Time Adaptation

arXiv:2609.35738 · cs.CL, cs.LG · Submitted 2026-09-28 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-09-28

Updated: 2026-09-28

Project page: https://schema-harness.github.io

Terminology

• Adaptation — 36× in this paper · explained in 1 episode
• Benchmark — 19× in this paper · explained in 2 episodes
• Reinforcement Learning — 14× in this paper · explained in 4 episodes
• Meta-learning — 12× in this paper · explained in 2 episodes
• Trace — 10× in this paper · explained in 2 episodes
• GRPO — 8× in this paper · explained in 6 episodes
• Fine-Tuning — 5× in this paper · explained in 5 episodes
• Continual Learning — 4× in this paper · explained in 2 episodes
• Generalization — 4× in this paper · explained in 13 episodes
• Scaffolding — 4× in this paper · explained in 3 episodes
• Verifiable Rewards — 3× in this paper · explained in 2 episodes
• Adam — 2× in this paper · explained in 2 episodes
• Agent Harness — 2× in this paper · explained in 2 episodes
• National Science Foundation — 2× in this paper
• World Model — 2× in this paper · explained in 7 episodes
• Agentic Coding — 1× in this paper · explained in 1 episode
• Alternation — 1× in this paper · explained in 1 episode
• Core — 1× in this paper · explained in 2 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Frozen model — 1× in this paper · explained in 3 episodes
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Self-Distillation — 1× in this paper · explained in 7 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning
• SIA: Self Improving AI with Harness & Weight Updates
• Meta-Learning in Neural Networks: A Survey
• Evo-Bench: Can Language Models Improve Agent Harness?
• Feedback Friction: LLMs Struggle to Fully Incorporate External Feedback
• WHALE: A Simple Recipe for Joint Harness-Weight Optimization
• Meta-Harness: End-to-End Optimization of Model Harnesses
• RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards
• Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
• EvoX: Meta-Evolution for Automated Discovery
• AutoHarness: improving LLM agents by automatically synthesizing a code harness
• TTHE: Test-Time Harness Evolution
• EvoHarness-RL: Learning Runtime Harness Coordination for Self-Evolving Agents
• The Last Harness You'll Ever Build
• Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards
• Training a Generally Curious Agent
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering