Papers › Simple, Good, Fast: Self-Supervised World Models Free of Baggage

Simple, Good, Fast: Self-Supervised World Models Free of Baggage

3 Jun 2025arXiv:2506.02612archive 2025-07-28

Jan Robine, Marc Höftmann, Stefan Harmeling

What are the essential components of world models? How far do we get with world models that are not employing RNNs, transformers, discrete representations, and image reconstructions? This paper introduces SGF, a Simple, Good, and Fast world model that uses self-supervised representation learning, captures short-time dependencies through frame and action stacking, and enhances robustness against model errors through data augmentation. We extensively discuss SGF's connections to established world models, evaluate the building blocks in ablation studies, and demonstrate good performance through quantitative comparisons on the Atari 100k benchmark.

PaperPDFCodeCode Syntology ranMethod as code

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 34 code samples harvested from 1 repository linked to this paper: 26 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 1 violated, 5 with no contract checked; 18 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 16 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 34 samples from 1 repository, 26 ran (8 with no instrument failure; 18 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 between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

jrobine/sgf officialmentioned in paperpytorchSyntology: 34 samples, 26 ran (18 instrument-failed) · 8 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

34 samples harvested: 26 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 1 violated, 5 with no contract checked; 18 where Syntology's instrument failed) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
1ran · violated contract
16ran · our draft was wrong
2ran; the contract check could not run
5ran
8unverified

Licence: 0 of the 34 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 jrobine/sgf. “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”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

jrobine/sgf official repository 34 samples, 26 ran (18 instrument-failed) · 8 unverified
Clip src/wm.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 0ef9795780017495 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
LayerNorm in 121 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
src/wm.py ran MIT (permissive) · 6ca16d3e5b87ea81 · report fixture: returned LayerNorm · level recorded 2026-09-22
Mask in 10 papers src/wm.py ran MIT (permissive) · 81c4e3c85779a38b · report fixture: returned Mask · level recorded 2026-09-22
Sign src/wm.py ran fingerprinted MIT (permissive) · 82abe86d62c22ea4 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
VectorMLP src/wm.py ran MIT (permissive) · 1802a24ad0e1881d · report fixture: returned VectorMLP · level recorded 2026-09-22
_get_factory src/wm.py ran · our draft was wrong MIT (permissive) · fe3e20372d4c072c · report fixture: signature factories:dict/0/?, value:str/0/str, key_id:str/0/str · returned partial · level recorded 2026-09-22
activation in 30 papers src/wm.py ran · our draft was wrong MIT (permissive) · afb336dc19e6c757 · report fixture: signature act:str/0/str, none_identity:bool/0/bool · returned ReLU · level recorded 2026-09-22
bins src/wm.py ran · honoured contract MIT (permissive) · 08417eca38375681 · report fixture: signature low:float/0/float, high:float/0/float, num:int/0/int, device:NoneType/0/? · returned Tensor · level recorded 2026-09-22
cnn src/wm.py ran · our draft was wrong MIT (permissive) · e616246bc18fd1b5 · report fixture: signature in_dim:int/0/int, in_res:int/0/int, out_dim:int/0/int, channels:str/0/str, kernels:str/0/str, strides:str/0/str, paddings:str/0/str, norm:str/0/str, act:str/0/str, init:str/0/str, out_bias:bool/0/bool, out_norm:str/0/str, memory_format:NoneType/0/?, device:NoneType/0/? · returned list · level recorded 2026-09-22
device in 8 papers src/wm.py ran · our draft was wrong MIT (permissive) · e332d55367824ddd · report fixture: signature mod:callable/0/? · returned device · level recorded 2026-09-22
get_mask in 119 papers
census: 9 implementations from 10 papers, 9 distinct outputs across 7 buckets, one shared input per bucket
src/wm.py ran · our draft was wrong MIT (permissive) · d31e9b2f069596f5 · report fixture: signature values:Tensor/2/bool · returned Mask · level recorded 2026-09-22
hinge in 6 papers src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · 15ba315adfdfafd6 · report fixture: signature x:Tensor/2/float, target:Tensor/2/float · returned Tensor · level recorded 2026-09-22
init_ in 12 papers src/wm.py ran · our draft was wrong MIT (permissive) · 29899c0203809b63 · report fixture: signature mod:callable/0/?, init:str/0/str · returned SimpleModule · level recorded 2026-09-22
map_structure src/wm.py ran · our draft was wrong MIT (permissive) · dce31dd97728c79d · report fixture: signature fn:callable/0/?, value:dict/0/?, skip_none:bool/0/bool · returned dict · level recorded 2026-09-22
mlp in 108 papers src/wm.py ran · our draft was wrong MIT (permissive) · 82d342964b41dc7f · report fixture: signature in_dim:int/0/int, dims:str/0/str, norm:str/0/str, act:str/0/str, init:str/0/str, out_bias:bool/0/bool, out_norm:bool/0/bool, device:NoneType/0/? · returned tuple · level recorded 2026-09-22
norm_1d src/wm.py ran · our draft was wrong MIT (permissive) · 4eb2fb2d744ed08b · report fixture: signature norm:str/0/str, d:int/0/int, device:NoneType/0/?, none_identity:bool/0/bool · returned LayerNorm · level recorded 2026-09-22
norm_2d src/wm.py ran · our draft was wrong MIT (permissive) · dd38f94f790d5f55 · report fixture: signature norm:str/0/str, d:int/0/int, device:NoneType/0/?, none_identity:bool/0/bool · returned LayerNorm · level recorded 2026-09-22
off_diagonal in 22 papers src/wm.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · cbcbe66ba5258fb1 · report fixture: signature x:Tensor/2/float, keepdim:bool/0/bool · returned Tensor · level recorded 2026-09-22
parse_dims src/wm.py ran · honoured contract MIT (permissive) · dfc0d2b811d785ae · report fixture: signature dims:str/0/str · returned tuple · level recorded 2026-09-22
sample_bernoulli in 9 papers src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · ab04fff3a9720e27 · report fixture: signature probs:Tensor/2/float · returned Tensor · level recorded 2026-09-22
sample_categorical in 20 papers src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · a8117d72551969e1 · report fixture: signature probs:Tensor/2/float · returned Tensor · level recorded 2026-09-22
symexp in 20 papers src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · 0d83ad2de737d8cb · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
symlog in 24 papers
census: 5 implementations from 7 papers, 3 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · d89a7360f2557af3 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
two_hot src/wm.py ran · our draft was wrong fingerprinted MIT (permissive) · 420038ff485d6c0b · report fixture: signature tensor:Tensor/2/float, bins:Tensor/1/float · returned Tensor · level recorded 2026-09-22
unsqueeze_as src/wm.py ran · violated contract fingerprinted MIT (permissive) · 2b187ef6125cd1bc · report fixture: signature tensor:Tensor/2/float, target_tensor:Tensor/4/float · returned Tensor · level recorded 2026-09-22
variance_covariance_loss src/wm.py ran; the contract check could not run (RAISES) MIT (permissive) · d8c67c4a6ad91b62 · report fixture: signature z:Tensor/2/float, contrastive:bool/0/bool, eps:float/0/float · returned tuple · level recorded 2026-09-22
ScalarMLP src/wm.py unverified MIT (permissive) · c15ab938fe840d9f · report
WorldModel in 13 papers src/wm.py unverified MIT (permissive) · 71323f43911a356d · report
_build_module src/wm.py unverified MIT (permissive) · 45bed2d814f991f3 · report
_norm_init_ src/wm.py unverified MIT (permissive) · 8bcb44beb386f1ab · report
normal_init_ src/wm.py unverified MIT (permissive) · 4216a828ef7176fb · report
space_flatten src/wm.py unverified MIT (permissive) · 148a74da6f2f20e6 · report
truncated_normal_init_ src/wm.py unverified MIT (permissive) · 100163aa698b77ce · report
zero_init_ src/wm.py unverified MIT (permissive) · 425012f1ad329944 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for SGF (graph method: SGF). 34 bricks; individually on a fixture: 26 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 1 violated, 5 with no contract checked; 18 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. WorldModel entry · level 0 · unverified · 71323f43911a356d
  2. ScalarMLP helper · level 0 · unverified · c15ab938fe840d9f
  3. VectorMLP helper · level 1 · ran · 1802a24ad0e1881d
  4. cnn model · level 1 · ran · our draft was wrong · e616246bc18fd1b5
  5. device helper · level 1 · ran · our draft was wrong · e332d55367824ddd
  6. get_mask helper · level 1 · ran · our draft was wrong · d31e9b2f069596f5
  7. space_flatten helper · level 0 · unverified · 148a74da6f2f20e6
  8. variance_covariance_loss helper · level 1 · ran; the contract check could not run (RAISES) · d8c67c4a6ad91b62
  9. bins helper · level 1 · ran · honoured contract · 08417eca38375681
  10. init_ helper · level 1 · ran · our draft was wrong · 29899c0203809b63
  11. mlp model · level 1 · ran · our draft was wrong · 82d342964b41dc7f
  12. sample_bernoulli helper · level 1 · ran · our draft was wrong · ab04fff3a9720e27
  13. sample_categorical helper · level 1 · ran · our draft was wrong · a8117d72551969e1
  14. symexp helper · level 1 · ran · our draft was wrong · 0d83ad2de737d8cb
  15. symlog helper · level 1 · ran · our draft was wrong · d89a7360f2557af3
  16. two_hot helper · level 1 · ran · our draft was wrong · 420038ff485d6c0b
  17. activation helper · level 1 · ran · our draft was wrong · afb336dc19e6c757
  18. norm_1d helper · level 1 · ran · our draft was wrong · 4eb2fb2d744ed08b
  19. norm_2d helper · level 1 · ran · our draft was wrong · dd38f94f790d5f55
  20. parse_dims helper · level 1 · ran · honoured contract · dfc0d2b811d785ae
  21. Mask helper · level 1 · ran · 81c4e3c85779a38b
  22. hinge helper · level 1 · ran · our draft was wrong · 15ba315adfdfafd6
  23. off_diagonal helper · level 1 · ran; the contract check could not run (RAISES) · cbcbe66ba5258fb1
  24. _get_factory helper · level 1 · ran · our draft was wrong · fe3e20372d4c072c
  25. _build_module helper · level 0 · unverified · 45bed2d814f991f3
  26. map_structure helper · level 1 · ran · our draft was wrong · dce31dd97728c79d
  27. unsqueeze_as helper · level 1 · ran · violated contract · 2b187ef6125cd1bc
  28. Clip helper · level 1 · ran · 0ef9795780017495
  29. LayerNorm helper · level 1 · ran · 6ca16d3e5b87ea81
  30. Sign helper · level 1 · ran · 82abe86d62c22ea4
  31. _norm_init_ helper · level 0 · unverified · 8bcb44beb386f1ab
  32. normal_init_ helper · level 0 · unverified · 4216a828ef7176fb
  33. truncated_normal_init_ helper · level 0 · unverified · 100163aa698b77ce
  34. zero_init_ helper · level 0 · unverified · 425012f1ad329944

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2506.02612") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Data AugmentationRepresentation 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