Pith Number

pith:TA7COXXJ

pith:2024:TA7COXXJ5JANU25JWM2F4OLAVR

not attested not anchored not stored refs resolved

The Platonic Representation Hypothesis

Brian Cheung, Minyoung Huh, Phillip Isola, Tongzhou Wang

Representations learned by different neural networks are converging toward a shared statistical model of reality.

arxiv:2405.07987 v5 · 2024-05-13 · cs.LG · cs.AI · cs.CV · cs.NE

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{TA7COXXJ5JANU25JWM2F4OLAVR}

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

We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality. We term such a representation the platonic representation.

C2weakest assumption

That observed similarities in representation spaces reflect convergence to an objective underlying model of reality rather than shared inductive biases, training data overlap, or architectural similarities.

C3one line summary

Representations learned by large AI models are converging toward a shared statistical model of reality.

References

272 extracted · 272 resolved · 38 Pith anchors

[1] Cognitive Systems Research , volume =

[2] Communications of the ACM , volume= 2017

[3] Nature communications , volume= 2018

[4] Text and Code Embeddings by Contrastive Pre-Training · arXiv:2201.10005

[5] Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

Formal links

2 machine-checked theorem links

Cited by

35 papers in Pith

A Systematic Evaluation of Co-folding Model Representations for Small-Molecule Learning

The physics of AI weather models

gen2seg: Generative Models Enable Generalizable Instance Segmentation

AI Consciousness and Existential Risk

Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers

Receipt and verification

First computed	2026-05-17T23:38:53.319583Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

983e275ee9ea40da6ba9b3345e3960ac4ac537627ad6c62795d551b0886d288a

Aliases

arxiv: 2405.07987 · arxiv_version: 2405.07987v5 · doi: 10.48550/arxiv.2405.07987 · pith_short_12: TA7COXXJ5JAN · pith_short_16: TA7COXXJ5JANU25J · pith_short_8: TA7COXXJ

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/TA7COXXJ5JANU25JWM2F4OLAVR \
  | 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: 983e275ee9ea40da6ba9b3345e3960ac4ac537627ad6c62795d551b0886d288a

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "7764d62685e75160e340bdbfad327e15116bcca1f61b1e7f9cea21f703b9bd0f",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CV",
      "cs.NE"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2024-05-13T17:58:30Z",
    "title_canon_sha256": "84711fac72d246673ea735c241943b761c2288440c296e17ea48e682c5019bba"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2405.07987",
    "kind": "arxiv",
    "version": 5
  }
}