The paper investigates whether restoring full sandbox snapshots can aid coding agents in recovering from failures. By implementing a Go‑Explore‑style search over complete Daytona snapshots, the authors find that higher and more efficient solve rates arise when progress is stored outside the repository—such as in databases, services, and other runtime state. The study suggests that full‑snapshot search is most beneficial for long‑horizon coding tasks that span multiple services whose state cannot be reconstructed from a simple diff.
arXiv:2608.29675v1 Announce Type: cross Abstract: Repository exploration is a distinct and costly stage of coding-agent pipelines: before generating a patch, an agent must identify which repository f...
Long-horizon coding requires agents to maintain an evolving view of the project they are changing: what problem is currently being solved, what has already been learned or modified, which other artifa...
Scaling coding agents requires a continuing supply of executable data for training, benchmarking, and continuous evaluation. Each task must couple a realistic software state with a specification, development tools, and reliable verification.
arXiv:2609.23570v1 Announce Type: cross Abstract: Coding agents operate on real repository coding tasks, and persistent memory systems promise to reuse experience across tasks. Yet existing evaluatio...
arXiv:2607. 28591v1 Announce Type: cross Abstract: Scaling coding agents requires a continuing supply of executable data for training, benchmarking, and continuous evaluation.
SWE-Shift is a new benchmark that tests coding agents in environments where tasks and workspaces can change mid-execution. It presents 120 scenarios across 12 change families, each evaluated at nine intervention points, comparing a Clean continuation (no change) with a Shifted continuation (task-relevant update). The benchmark shows that all evaluated agents perform better on Clean than on Shifted, with significant drops in success rates and zero paired success for certain change types.
arXiv:2607. 22711v1 Announce Type: cross Abstract: LLM coding agents operate by constructing trajectories that accumulate reasoning, tool calls, and results to enable multi-step decision-making.
arXiv:2607. 01916v1 Announce Type: new Abstract: Large language model agents can repair real repository issues, but they often spend large context budgets on whole-file reads, broad searches, and long terminal outputs where useful evidence is mixed with irrelevant code and logs.
arXiv:2606. 29823v1 Announce Type: cross Abstract: The database community has repeatedly advanced the state of the art by recognizing that new workloads demand new system architectures.
arXiv:2607. 18161v1 Announce Type: cross Abstract: Coding agents are increasingly used to accelerate code generation in many downstream tasks, such as fixing bugs, building applications, and prototyping.
arXiv:2606.22906v2 Announce Type: replace-cross Abstract: Large language models have shown strong performance on software engineering (SE) tasks, yet understanding large industrial repositories remai...
arXiv:2606. 11976v1 Announce Type: cross Abstract: Software engineering tools increasingly rely on LLM based agents to localize files to change to resolve a software issue.