Learning and Transferring Closed-Loop Robot Software

Your coding assistant makes changes that still need substantial repair before merging.

What you would change

Try the technique on held-out issues from a repository with reliable tests.

Try the technique on held-out issues from a repository with reliable tests.

Compare against

The same coding agent, repository snapshot, tools, and token budget.

1. Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.
2. Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.
3. Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.
4. Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test.

What to measure

Correct patches, regression failures, human repair time, and total cost per accepted change.

When to keep it

Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation.

recommendation

test

research area

coding agents

technical change

Archived execution-optimized policy code for cross-task transfer

published

2026-09-17

The method is feasible on constrained infrastructure since it needs only third-party coding-agent APIs, an open-source simulator, and no model training, but the transfer benefit is task-dependent and validated only in simulation at moderate success rates, so it justifies a workload-specific small-scale trial rather than direct production adoption.

1. claim 01Paperraft inference
   Read the brief interpretation

   Replaces manual design and tuning of closed-loop robot policy code (observation processing, state management, situation-dependent branching) with coding-agent generation that is iteratively refined by simulation feedback, archived, and reused as reference implementations when acquiring policies for new tasks. Costs consist of repeated coding-agent API calls and simulation rollouts during optimization and archive construction, plus the need for task demonstrations and a relevant simulator; deployment itself is inexpensive because the selected policy is frozen and executes without further model calls. Transfer is inconsistent (initial unoptimized references were better on two of nine target tasks), all evidence is simulation-only so the sim-to-real gap is unquantified, and the 57% mean success rate is below the reliability most production settings require.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source
2. claim 02Paperraft inference

   Iterative simulation-feedback optimization raises mean source-task success from 28.3% to 64.2%; across nine target tasks, optimized archived references yield 57.0% mean success versus 45.2% without references and 41.5% with unoptimized initial code, a mean gain of 15.6 percentage points over initial references.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source