Pith Number

pith:5DCOF3VM

pith:2021:5DCOF3VMGTZLMLR7G3WPGXB35H

not attested not anchored not stored refs resolved

Emerging Properties in Self-Supervised Vision Transformers

Armand Joulin, Herv\'e J\'egou, Hugo Touvron, Ishan Misra, Julien Mairal, Mathilde Caron, Piotr Bojanowski

Self-supervised Vision Transformers encode explicit semantic segmentation information in their features.

arxiv:2104.14294 v2 · 2021-04-29 · cs.CV

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{5DCOF3VMGTZLMLR7G3WPGXB35H}

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

self-supervised ViT features contain explicit information about the semantic segmentation of an image, which does not emerge as clearly with supervised ViTs, nor with convnets [...] achieving 80.1% top-1 on ImageNet in linear evaluation with ViT-Base.

C2weakest assumption

The assumption that the observed semantic segmentation information and k-NN performance arise specifically from the interaction of self-supervision with the ViT architecture rather than from particular hyperparameter choices, dataset statistics, or evaluation protocols.

C3one line summary

Self-supervised ViTs show emergent semantic segmentation and 78.3% k-NN accuracy on ImageNet; DINO reaches 80.1% linear evaluation with ViT-Base.

References

85 extracted · 85 resolved · 17 Pith anchors

[1] arXiv preprint arXiv:1804.03235 , year= 2018

[2] Self-labelling via simultaneous clustering and repre- sentation learning 2020

[3] preprint arXiv:2006.10803 , year= 2006

[4] Neural Machine Translation by Jointly Learning to Align and Translate 2014 · arXiv:1409.0473

[5] MultiGrain : a unified image embedding for classes and instances 1902 · arXiv:1902.05509

Formal links

2 machine-checked theorem links

Cited by

34 papers in Pith

CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology

Wan: Open and Advanced Large-Scale Video Generative Models

VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction

TextTeacher: What Can Language Teach About Images?

SSL4RL: Revisiting Self-supervised Learning as Intrinsic Reward for Visual-Language Reasoning

Receipt and verification

First computed	2026-05-17T23:38:47.666608Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

e8c4e2eeac34f2b62e3f36ecf35c3be9dba589951a95e029295327006e6a9849

Aliases

arxiv: 2104.14294 · arxiv_version: 2104.14294v2 · doi: 10.48550/arxiv.2104.14294 · pith_short_12: 5DCOF3VMGTZL · pith_short_16: 5DCOF3VMGTZLMLR7 · pith_short_8: 5DCOF3VM

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/5DCOF3VMGTZLMLR7G3WPGXB35H \
  | 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: e8c4e2eeac34f2b62e3f36ecf35c3be9dba589951a95e029295327006e6a9849

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "e27e3a9f2795b58bc613cf7732e789f220f1f340ac0c88e34c09aebf18ca1c51",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2021-04-29T12:28:51Z",
    "title_canon_sha256": "6da5120c2f200ed38479654dc470ad04e74b90b6056c9f2193cbebf59211b2ef"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2104.14294",
    "kind": "arxiv",
    "version": 2
  }
}