SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.31111 · 2026

Subspace-Decomposed JEPAs: Disentangling Progression and Content in Latent World Models

Jesse Read, Lucas Thil, École Palaiseau, Guillaume Doquet, D In O -W M + P Ro P P L D M D In O -W, M + P Ro P P L D M D In O -W M G C B C, ) L E W M P L D M D In O -W

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2605.31111.json · badge: /api/ran/2605.31111.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
LucasStill/SD-JEPA community
Papers with Code archive
5 ran · 1 no recorded run
FunctionStatusWhere it lives
canonical_injection Ran
ran, honoured the contract we drafted
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("c8fedbbcb0a259d7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

canonical_injection_complement Ran
ran, honoured the contract we drafted
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("b2968fcedd512aba")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

canonical_prototypes Ran
ran, honoured the contract we drafted
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("15c931acb1cb0262")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

detach_clone Ran
ran, honoured the contract we drafted
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("5ba227e01dd31840")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

polar_features Ran
ran; our contract draft was wrong, not the code
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("88402b13ad89fbad")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

JEPA Not yet run
unverified (harvested, no recorded run)
LucasStill/SD-JEPA/le-wm-step/jepa.py
pointer only (licence: NONE) · get_code("69c266e3a28a14f2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LucasStill/SD-JEPA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Joint-Embedding Predictive Architectures (JEPAs) learn compact latent world models by predicting future embeddings, but no single coordinate of the latent is designated to encode task progression. We carve the JEPA latent into two orthogonal subspaces with disjoint roles: a low-dimensional progression subspace z prog ∈ R k shaped by a cosine-margin triplet loss, and a high-dimensional content subspace z cont ∈ R D-k regularised by the existing SIGReg objective of LEWM. We prove that the two anti-collapse forces act on disjoint coordinates, so they compose additively rather than competing on the same dimensions. Our method, SD-JEPA improves over the LEWM baseline on the majority of its control benchmarks at matched compute, and outperforms the strongest non-LEWM JEPA baseline on Push-T; a subspace-ablation falsifier confirms the split is the load-bearing ingredient. Beyond planning, the resulting 1-D angular progression coordinate θ t = atan2(z prog 2 , z prog 1 ) functions as a scene-aware compass on the latent. It advances with task progress, regresses when the agent backtracks, and under controlled perturbations both spikes and relocalises to a semantically appropriate new task-phase sector, separating the moment of surprise from its meaning in a way that prediction-error scalars cannot. Three quantitative tests back this up: |∆θ t | outperforms the standard latent-prediction-error surprise (z-MSE) at localising semantic events on 40 held-out cube episodes by up to +0.18 pooled AUROC (97.5% perepisode win rate at ±1-step tolerance); a within-episode linear probe across all four environments (40 episodes per env) shows the 8-dimensional progression subspace (4.2% of the latent) explains 72-95% of task-progress variance, with the largest probe-vs-clock gap on Reacher tracking the only env with a robust z prog -aware planning lift. Code at https://github.com/LucasStill/SD-JEPA Preprint.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2605.31111")
get_code_for_paper("2605.31111")
have("2605.31111")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2605.31111.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2605.31111.svg)](https://syntology.ai/paper/2605.31111)

Connect an agent — have() is free.