Pith Number

pith:CGB34AKJ

pith:2026:CGB34AKJUCDLA4MRT5AZDRXZRR

not attested not anchored not stored refs pending

jina-embeddings-v5-omni: Geometry-preserving Embeddings via Locked Aligned Towers

Andreas Koukounas, Florian H\"onicke, Han Xiao, Michael G\"unther, Mohammad Kalim Akram, Saba Sturua, Scott Martens

GELATO extends existing text embedding models to images, audio and video by freezing nearly all weights and training only the connectors.

arxiv:2605.08384 v3 · 2026-05-08 · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{CGB34AKJUCDLA4MRT5AZDRXZRR}

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

Our evaluations show that GELATO produces results that are competitive with the state-of-the-art, yielding nearly equal performance to larger multimodal embedding models.

C2weakest assumption

The assumption that freezing the backbone text embedding models and non-text modality encoders while training only the connecting components will preserve semantic geometry and enable effective cross-modal alignment without degrading original text performance.

C3one line summary

GELATO extends frozen text embedding models with locked image and audio encoders, training minimal connectors to produce a single semantic embedding space for text, image, audio, and video while keeping original text performance unchanged.

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models

Receipt and verification

First computed	2026-06-09T01:04:43.448778Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

1183be0149a086b071919f4191c6f98c6c79a0ad26d4379f806c5e71967b0f24

Aliases

arxiv: 2605.08384 · arxiv_version: 2605.08384v3 · doi: 10.48550/arxiv.2605.08384 · pith_short_12: CGB34AKJUCDL · pith_short_16: CGB34AKJUCDLA4MR · pith_short_8: CGB34AKJ

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/CGB34AKJUCDLA4MRT5AZDRXZRR \
  | 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: 1183be0149a086b071919f4191c6f98c6c79a0ad26d4379f806c5e71967b0f24

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "ea6470d86e28138f1baf8a6d4f9cb1cd4938bdb7391a816f931cc60580c14b78",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-05-08T18:45:15Z",
    "title_canon_sha256": "9540f9bd214d9de24038e1f3e6692af44f4e1d34e099310e4337c05a2bf1996e"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.08384",
    "kind": "arxiv",
    "version": 3
  }
}