Pith Number

pith:OA7KLRV4

pith:2026:OA7KLRV4HOEZ3NMINSQB2WBOSX

not attested not anchored not stored refs resolved

LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

Damien Scieur, Lucas Maes, Quentin Le Lidec, Randall Balestriero, Yann LeCun

LeWorldModel trains the first stable end-to-end JEPA from raw pixels using only two loss terms.

arxiv:2603.19312 v2 · 2026-03-13 · cs.LG · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{OA7KLRV4HOEZ3NMINSQB2WBOSX}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

LeWM is the first JEPA that trains stably end-to-end from raw pixels using only two loss terms: a next-embedding prediction loss and a regularizer enforcing Gaussian-distributed latent embeddings, reducing tunable loss hyperparameters from six to one.

C2weakest assumption

That the Gaussian regularizer alone is sufficient to prevent representation collapse across diverse 2D and 3D control tasks without any auxiliary supervision or pre-trained encoders.

C3one line summary

LeWM is the first end-to-end trainable JEPA from pixels that uses only two loss terms for stable training and fast planning on 2D/3D control tasks.

References

57 extracted · 57 resolved · 13 Pith anchors

[1] End-to-end training of deep visuomotor policies.Journal of Machine Learning Research, 17(39):1–40, 2016 2016

[2] World Models 2018 · arXiv:1803.10122

[3] Transformers are sample-efficient world models 2023

[4] Training Agents Inside of Scalable World Models 2025 · arXiv:2509.24527

[5] A path towards autonomous machine intelligence version 0.9 2022

Formal links

3 machine-checked theorem links

Cited by

28 papers in Pith

IntentScore: Intent-Conditioned Action Evaluation for Computer-Use Agents

LaMo: Self-Supervised Latent Motion Priors for Physical Realism in Video Generation

Pelican-Unify 1.0: A Unified Embodied Intelligence Model for Understanding, Reasoning, Imagination and Action

stable-worldmodel: A Platform for Reproducible World Modeling Research and Evaluation

Beyond Euclidean Proximity: Repairing Latent World Models with Horizon-Matched Trajectory Reachability Metrics

Receipt and verification

First computed	2026-05-17T23:38:53.584673Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

703ea5c6bc3b899db5886ca01d582e95d47de7674dcd3215a243ba77cc8114f7

Aliases

arxiv: 2603.19312 · arxiv_version: 2603.19312v2 · doi: 10.48550/arxiv.2603.19312 · pith_short_12: OA7KLRV4HOEZ · pith_short_16: OA7KLRV4HOEZ3NMI · pith_short_8: OA7KLRV4

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/OA7KLRV4HOEZ3NMINSQB2WBOSX \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 703ea5c6bc3b899db5886ca01d582e95d47de7674dcd3215a243ba77cc8114f7

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "6a9f8b9150e3eed0763657a4c826c7ce16de4fc43189b9aea2b3d38f9c9389c8",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-03-13T19:48:14Z",
    "title_canon_sha256": "c996afe7bd20ae820ffebacb3d8545c079f16c9dbec5cd86e4964808eaa7e042"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2603.19312",
    "kind": "arxiv",
    "version": 2
  }
}