Persistent Recursive Worlds Enable Autonomous Software Evolution
- URL: http://arxiv.org/abs/2608.10450v3
- Date: Sun, 16 Aug 2026 07:53:00 GMT
- Title: Persistent Recursive Worlds Enable Autonomous Software Evolution
- Abstract summary: EvoX Genesis makes software project persistent while allowing local agents to remain finite-lived.<n>We evaluate this organization across formation, continuation and redevelopment.<n>Results show that long-horizon software development can be organized around a persistent project rather than a persistent agent.
- Score: 9.245787953561281
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Complex software systems develop over timescales that exceed the lifespan of any individual coding agent. Most agentic software systems preserve continuity through persistent sessions, memories, managers or shared context. We introduce EvoX Genesis (hereafter, Genesis), which instead makes the software project persistent while allowing local agents to remain finite-lived. Genesis represents software as a persistent recursive world: each local world is situated by an accepted version and a repository path, finite-lived agents propose local changes, recursive delegation moves work across paths, and only accepted consequences advance the persistent version history. We evaluate this organization across formation, continuation and redevelopment. Starting from a repository with no compiler implementation, Genesis used DeepSeek V4 Flash to build a Rust-based C compiler with about 250k tracked lines; the run lasted over 120 hours, archived over 1,000 agent episodes and incurred only US\$44 in model-token charges. The compiler passed the complete c-testsuite and most LLVM and Csmith tests. In a separate compiler world generated with GLM 5.2, development continued after repeated agent replacement while retaining full test performance. Genesis also reimplemented 13 MESA modules with over 100k Fortran lines as a Rust workspace with nearly 90k Rust lines; across six numerical workloads, it achieved median speedups of 1.55--6.87x. These results show that long-horizon software development can be organized around a persistent project rather than a persistent agent. Project Website: https://genesis.evox.group/
Related papers
- Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments [89.86581796261123]
We introduce Terminal-Universe, a framework which turns each trajectory into a reusable environment.<n>Terminal-Universe replays the file operations recorded in a trajectory to restore each file before the agent modified it, yielding a partial workspace.<n>We also scale the tasks along two axes: breadth and depth.
arXiv Detail & Related papers (2026-09-03T17:41:05Z) - LoopsBench: From Harness Engineering to Loop Engineering in Benchmarking Coding Agent [60.27239351179265]
LOOPSBENCH is a benchmark for loop engineering in coding agent evaluation.<n>It comprises 112 tasks from authentic sources spanning 8 programming languages and 9 domains.<n>Its flow-aware runtime releases tests along the ready frontier and retains completed nodes as regression obligations.
arXiv Detail & Related papers (2026-07-31T20:18:25Z) - An Empirical Study of Coordination Mode as the First-Class Citizen in From-Scratch Multi-Agent Coding [9.678180803829207]
We introduce multi-agent evaluation benchmark, MSEval, evaluating multi-agent coding on real-world tasks.<n>Grounded in 10 authentic, full-stack projects across 10 domains, MSEval scores performance using hierarchical requirements and deterministic rubrics.
arXiv Detail & Related papers (2026-07-30T08:53:05Z) - Trustworthy Software Project Generation : a Case Study with an Interactive Theorem Prover [2.6146136791093104]
This paper studies whether interactive theorem provers (ITPs) can support large-scale software generation.<n>Because ITPs handle pure total functions but not effects such as I/O, our agent separates effectful code from pure logic.<n>We study this route through the fully automatic development of a CPU interpreter for all 47 instructions of the unprivileged RISC-V RV32I base.
arXiv Detail & Related papers (2026-05-25T16:35:36Z) - RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades [32.04167306620878]
We present RoadmapBench, a benchmark of 115 long-horizon coding tasks grounded in real open-source version upgrades.<n>Each task places the agent on a source-version code snapshot and provides a multi-target roadmap instruction.<n>We find that even the strongest, Claude-Opus-4.7, resolves only 39.1% of tasks, while the weakest iteration achieves merely 5.2%.
arXiv Detail & Related papers (2026-05-15T11:00:33Z) - KISS Sorcar: A Stupidly-Simple General-Purpose and Software Engineering AI Assistant [8.273738980475393]
KISS Sorcar is a general-purpose assistant and integrated development environment built on top of the KISS Agent Framework.<n> KISS Sorcar is a free, open-source Visual Studio Code extension that runs locally and effectively for long-horizon tasks.<n>The entire system was built using itself in 4.5 months, providing a continuous stress test in which any agent-introduced bug immediately impairs its own ability to work.
arXiv Detail & Related papers (2026-04-26T17:59:06Z) - ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents [77.22389710754452]
We introduce bench, a benchmark for coworker agents built around multi-turn multi-day tasks.<n>The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services.<n>The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0%, indicating that partial progress is common.
arXiv Detail & Related papers (2026-04-26T16:05:02Z) - SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks [55.76734816061826]
We introduce SlopCodeBench, a language-agnostic benchmark comprising 20 problems and 93 checkpoints.<n>We track two trajectory-level quality signals: verbosity, the fraction of redundant or duplicated code, and structural erosion.<n>No agent solves any problem end-to-end across 11 models; the highest checkpoint solve rate is 17.2%.
arXiv Detail & Related papers (2026-03-25T19:26:44Z) - NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents [79.29376673236142]
Existing benchmarks fail to rigorously evaluate the long-horizon capabilities required to build complete software systems.<n>We present NL2Repo Bench, a benchmark explicitly designed to evaluate the long-horizon repository generation ability of coding agents.
arXiv Detail & Related papers (2025-12-14T15:12:13Z) - Git Context Controller: Manage the Context of LLM-based Agents like Git [6.521644491529639]
Large language model (LLM) based agents have shown impressive capabilities by interleaving internal reasoning with external tool use.<n>We introduce Git-Context-Controller (GCC), a structured context management framework inspired by software version control systems.<n>In a self-replication case study, a GCC-augmented agent builds a new CLI agent from scratch, achieving 40.7 task resolution, compared to only 11.7 without GCC.
arXiv Detail & Related papers (2025-07-30T08:01:45Z) - SwingArena: Competitive Programming Arena for Long-context GitHub Issue Solving [90.32201622392137]
We present SwingArena, a competitive evaluation framework for Large Language Models (LLMs)<n>Unlike traditional static benchmarks, SwingArena models the collaborative process of software by pairing LLMs as iterations, who generate patches, and reviewers, who create test cases and verify the patches through continuous integration (CI) pipelines.
arXiv Detail & Related papers (2025-05-29T18:28:02Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.