analiza aiqbench
This arXiv preprint proposes SpecBuild, an LLM agent that speeds up repository environment setup by running alternative setup routes and predicted repair branches in parallel and keeping the first one that succeeds. On the authors' Multi-Docker-Eval benchmark (334 tasks, 40 repositories, 9 languages), they report 6.4 to 12.0 point gains in success rate and 22.4% to 41.6% lower setup time across three backbone models. These are self-reported results on one benchmark, so readers should check how well they hold up on other environments and what the extra parallel compute costs.
Speculative Forking: Turning Idle Wait into Parallel Exploration for Repository Environment Setup
De la arXiv
Text integralarXiv:2610.33129v1 Announce Type: cross Abstract: Before developers can use a software project, they must first get it to run. This often requires installing the right language runtime, system packages, build tools, and test commands. Recent LLM-based agents automate environment setup by inspecting a repository, writing setup scripts, running them, and repairing the resulting errors. However, these agents still follow a sequential trial-and-error loop: they try one setup route, wait for slow commands to finish, observe the failure, and only then decide what to try next.
Because environment commands often take long to run, this sequential workflow makes setup slow and limits exploration to one route at a time. We present SpecBuild, an LLM-based environment setup system designed to be both faster and more successful through speculative forking. SpecBuild explores candidate setup branches in parallel, returns the first committed environment, and discards the rest. It applies this design at two granularities: strategy-level forking explores alternative setup routes, such as different base images and dependency policies, while command-level forking predicts failures of running commands and prepares repair branches before those failures are observed.
We evaluate SpecBuild on Multi-Docker-Eval, covering 334 tasks from 40 repositories and 9 programming languages. Across three backbone models, SpecBuild improves the success rate by 6.4 to 12.0 percentage points over the strongest baseline, while reducing average setup time by 22.4% to 41.6%. These results show that LLM-based environment setup need not trade speed for reliability: parallel speculative exploration can make agents both faster and more successful.
Abstract:Before developers can use a software project, they must first get it to run. This often requires installing the right language runtime, system packages, build tools, and test commands. Recent LLM-based agents automate environment setup by inspecting a repository, writing setup scripts, running them, and repairing the resulting errors. However, these agents still follow a sequential trial-and-error loop: they try one setup route, wait for slow commands to finish, observe the failure, and only then decide what to try next.
Because environment commands often take long to run, this sequential workflow makes setup slow and limits exploration to one route at a time. We present SpecBuild, an LLM-based environment setup system designed to be both faster and more successful through speculative forking. SpecBuild explores candidate setup branches in parallel, returns the first committed environment, and discards the rest. It applies this design at two granularities: strategy-level forking explores alternative setup routes, such as different base images and dependency policies, while command-level forking predicts failures of running commands and prepares repair branches before those failures are observed.
We evaluate SpecBuild on Multi-Docker-Eval, covering 334 tasks from 40 repositories and 9 programming languages. Across three backbone models, SpecBuild improves the success rate by 6.4 to 12.0 percentage points over the strongest baseline, while reducing average setup time by 22.4% to 41.6%. These results show that LLM-based environment setup need not trade speed for reliability: parallel speculative exploration can make agents both faster and more successful.
Software Engineering (cs.SE); Computation and Language (cs.CL); Machine Learning (cs.LG)
arXiv:2610.33129 [cs.SE]
(or arXiv:2610.33129v1 [cs.SE] for this version)
https://doi.org/10.48550/arXiv.2610.33129 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 342 cuvinte
Subiecte conexe
TEAM: Temporal-Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
arXiv · 4 aug. 2026
Parason: Revealing Subtask and Trial Parallelism in LLM Reasoning
arXiv · 27 aug. 2026
When Drafts Evolve: Speculative Decoding Meets Online Learning
arXiv · 7 aug. 2026
Speculative Diffusion is a Tale of Two Tradeoffs
arXiv · 27 oct. 2026