# GDPevo: Evaluating Agent Self-Evolution on Real Business Tasks

arXiv: https://arxiv.org/abs/2608.03764 · HTML: https://mlsanity.com/paper/2608.03764/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 37 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Builds on (most cited first, 30 of 30)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [ExpeL: LLM Agents Are Experiential Learners](https://mlsanity.com/paper/2308.10144/) (2023)
- [A Comprehensive Survey of Continual Learning: Theory, Method and Application](https://mlsanity.com/paper/2302.00487/) (2023)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces](https://mlsanity.com/paper/2601.11868/) (2026)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [LiveBench: A Challenging, Contamination-Limited LLM Benchmark](https://mlsanity.com/paper/2406.19314/) (2024)
- [A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence](https://mlsanity.com/paper/2507.21046/) (2025)
- [GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks](https://mlsanity.com/paper/2510.04374/) (2025)
- [InfiAgent-DABench: Evaluating Agents on Data Analysis Tasks](https://mlsanity.com/paper/2401.05507/) (2024)
- [SpreadsheetBench: Towards Challenging Real World Spreadsheet Manipulation](https://mlsanity.com/paper/2406.14991/) (2024)
- [WorkBench: a Benchmark Dataset for Agents in a Realistic Workplace Setting](https://mlsanity.com/paper/2405.00823/) (2024)
- [AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge](https://mlsanity.com/paper/2412.13670/) (2024)
- [Reward Is Enough: LLMs Are In-Context Reinforcement Learners](https://mlsanity.com/paper/2506.06303/) (2025)
- [SkillFlow:Benchmarking Lifelong Skill Discovery and Evolution for Autonomous Agents](https://mlsanity.com/paper/2604.17308/) (2026)
- [SOP-Bench: Complex Industrial SOPs for Evaluating LLM Agents](https://mlsanity.com/paper/2506.08119/) (2025)
- [FHIR-AgentBench: Benchmarking LLM Agents for Realistic Interoperable EHR Question Answering](https://mlsanity.com/paper/2509.19319/) (2025)
- [EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments](https://mlsanity.com/paper/2607.05155/) (2026)
- [RSIBench-Data: Benchmarking Data-Centric Research for Recursive Self-Improvement](https://mlsanity.com/paper/2607.25886/) (2026)
- [BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation for Large Language Models via Lens of Dynamic Interactions](https://mlsanity.com/paper/2510.05318/) (2025)
- [SEA-Eval: A Benchmark for Evaluating Self-Evolving Agents Beyond Episodic Assessment](https://mlsanity.com/paper/2604.08988/) (2026)
- [EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer](https://mlsanity.com/paper/2607.05202/) (2026)
- [JobBench: Aligning Agent Work With Human Will](https://mlsanity.com/paper/2605.26329/) (2026)
- [CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?](https://mlsanity.com/paper/2605.16679/) (2026)
- [LLMigrate: Transforming "Lazy" Large Language Models into Efficient Source Code Migrators](https://mlsanity.com/paper/2503.23791/) (2025)
- [SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows?](https://mlsanity.com/paper/2605.15777/) (2026)
- [BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents](https://mlsanity.com/paper/2605.29225/) (2026)

## Built on by (most cited first, 3 of 3)

- [TTSE: A Two-Track Online Self-Evolution Framework](https://mlsanity.com/paper/2609.24289/) (2026)
- [From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement](https://mlsanity.com/paper/2610.14625/) (2026)
- [MetaExp: Learning to Self-Improve through Experience via Verifiable Cross-Task Environment Scaling](https://mlsanity.com/paper/2610.21492/) (2026)

## Abstract (arXiv)

Agent self-evolution updates an agent's persistent state from prior experience and reuses it to solve related tasks more effectively. Evaluating self-evolution is difficult: existing benchmarks provide limited coverage of economically valuable task domains, do not always design training and test tasks such that test-time gains can be attributed to training experience, and remain vulnerable to data contamination. We present GDPevo, an evolution-native benchmark grounded in GDP-related enterprise workflows, together with the fully automated data pipeline that generates it. Its core mechanism, rule hybridization, decomposes each enterprise workflow into atomic business rules, distributes subsets of these rules across training tasks, and recombines them in held-out test tasks so that test-time gains are attributable. GDPevo spans CRM, ERP, finance, healthcare, legal, and data-centric workflows. Its V1 release contains 120 tasks in 12 groups, with five training and five held-out test tasks per group. Full automation enables the pipeline to expand the suite to 240 tasks in 24 groups (V2) within two days, providing a practical response to contamination. Using GDPevo, we evaluate four agents, each comprising a harness and a model, under four supervision types. Self-evolution consistently improves held-out accuracy by up to 16.44 percentage points. But the best evolved agents remain far below the fully informed oracle ceiling of 91.6%, indicating that the self-evolution ability of current agents remains far from fully realized. We publicly release the pipeline, benchmark, and full evaluation results at https://github.com/Prism-Shadow/GDPevo.