Papers › Learning to Edit Visual Programs with Self-Supervision

Learning to Edit Visual Programs with Self-Supervision

4 Jun 2024arXiv:2406.02383archive 2025-07-28

R. Kenny Jones, Renhao Zhang, Aditya Ganeshan, Daniel Ritchie

We design a system that learns how to edit visual programs. Our edit network consumes a complete input program and a visual target. From this input, we task our network with predicting a local edit operation that could be applied to the input program to improve its similarity to the target. In order to apply this scheme for domains that lack program annotations, we develop a self-supervised learning approach that integrates this edit network into a bootstrapped finetuning loop along with a network that predicts entire programs in one-shot. Our joint finetuning scheme, when coupled with an inference procedure that initializes a population from the one-shot model and evolves members of this population with the edit network, helps to infer more accurate visual programs. Over multiple domains, we experimentally compare our method against the alternative of using only the one-shot model, and find that even under equal search-time budgets, our editing-based paradigm provides significant advantages.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 12 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 7 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 12 samples from 1 repository, 12 ran (5 with no instrument failure; 7 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 3 of the 12 samples that ran have no recorded date.

rkjones4/vpi-edit officialmentioned in paperpytorchNOASSERTIONSyntology: 12 samples, 12 ran (7 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

12 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 7 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

7ran · our draft was wrong
5ran

Licence: 12 of the 12 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from rkjones4/vpi-edit. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

rkjones4/vpi-edit official repository 12 samples, 12 ran (7 instrument-failed) · 0 unverified
getArgs utils.py ran · our draft was wrong licence not identified · pointer only · cdd558f65a47c91b · report fixture: signature arg_list:list/1/tuple · returned Namespace · level recorded 2026-09-24
get_init_population joint_infer.py ran · our draft was wrong licence not identified · pointer only · 37318229cc619a07 · report fixture: signature os_inf_net:callable/0/?, domain:Domain/0/?, pop_size:int/0/int, target:Tensor/2/float · returned list · level recorded, date not recorded
get_transform_embs edit_models.py ran · our draft was wrong licence not identified · pointer only · c13cbbc423c4826c · report fixture: signature net:callable/0/?, codes:Tensor/3/float · returned Tensor · level recorded, date not recorded
load_2d_vis_encoder model_utils.py ran · our draft was wrong licence not identified · pointer only · 35a71bc05cae070a · report fixture: signature domain:MockDomain/0/? · returned V2DCNN · level recorded 2026-09-24
load_3d_vis_encoder model_utils.py ran · our draft was wrong licence not identified · pointer only · 4eb432c4c691eecf · report fixture: signature domain:MockDomain/0/? · returned V3DCNN · level recorded 2026-09-24
load_vis_encoder model_utils.py ran · our draft was wrong licence not identified · pointer only · 95e865702c74dca6 · report fixture: signature domain:MockDomain/0/? · returned V3DCNN · level recorded 2026-09-24
make_oneshot_population joint_infer.py ran · our draft was wrong licence not identified · pointer only · b158b355e872bd21 · report fixture: signature os_inf_net:callable/0/?, domain:StandInDomain/0/?, pop_size:int/0/int, target:Tensor/2/float · returned list · level recorded, date not recorded
mergeArgs utils.py ran licence not identified · pointer only · 2356aa163d6b705f · report fixture: signature spec:list/1/tuple, gen:list/1/tuple · returned list · level recorded 2026-09-24
model_train train_utils.py ran licence not identified · pointer only · 120a58a5d94dcb27 · report fixture: signature loader:SimpleDataLoader/0/?, net:callable/0/?, opt:SimpleOptimizer/0/? · returned dict · level recorded 2026-09-24
model_train_batch train_utils.py ran licence not identified · pointer only · 6a1dbc6db4557ba6 · report fixture: signature batch:tuple/1/Tensor, net:callable/0/?, opt:SGD/0/? · returned dict · level recorded 2026-09-24
os_inference infer_progs.py ran licence not identified · pointer only · e35e76a3dcb69b0d · report fixture: signature os_inf_net:callable/0/?, args:ArgsConfig/0/?, vinput:list/1/list, ret_extra_info:bool/0/bool · returned tuple · level recorded 2026-09-24
print_results utils.py ran licence not identified · pointer only · 90e3f640250efb11 · report fixture: signature LOG_INFO:list/1/tuple, result:dict/0/?, args:Args/0/?, label:str/0/str, ret_early:bool/0/bool · returned dict · level recorded 2026-09-24

Tasks

Self-Supervised Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections