JET: Judge-Guided Evolution at Test Time for Agent Programs
cs.SE, cs.AI
Submitted: 2026-09-28
Updated: 2026-09-28
Project page: https://trinkle23897.github.io/learning-beyond-gradients
Sources
- Learning Robust Rewards with Adversarial Inverse Reinforcement Learning
- Quantifying Differences in Reward Functions
- Automated Design of Agentic Systems
- DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
- Meta-Harness: End-to-End Optimization of Model Harnesses
- Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark
- TTHE: Test-Time Harness Evolution
- AlphaEvolve: A coding agent for scientific and algorithmic discovery
- ReAct: Synergizing Reasoning and Acting in Language Models
- JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolution
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties