Pith Number

pith:X5N46CMJ

pith:2023:X5N46CMJETB4TB5ED6TJANC2FV

not attested not anchored not stored refs resolved

Sigmoid Loss for Language Image Pre-Training

Alexander Kolesnikov, Basil Mustafa, Lucas Beyer, Xiaohua Zhai

A pairwise sigmoid loss for image-text pre-training achieves 84.5% zero-shot ImageNet accuracy using only four TPU chips in two days.

arxiv:2303.15343 v4 · 2023-03-27 · cs.CV · 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{X5N46CMJETB4TB5ED6TJANC2FV}

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

Combined with Locked-image Tuning, with only four TPUv4 chips, we train a SigLiT model that achieves 84.5% ImageNet zero-shot accuracy in two days.

C2weakest assumption

That the sigmoid loss, which forgoes global batch normalization, will continue to produce high-quality representations when scaled to new datasets or model sizes without additional hyper-parameter tuning.

C3one line summary

SigLIP replaces softmax-based contrastive loss with a simple pairwise sigmoid loss for vision-language pre-training, decoupling batch size from normalization and reaching strong zero-shot performance with limited compute.

References

60 extracted · 60 resolved · 9 Pith anchors

[1] Getting vit in shape: Scaling laws for compute-optimal model design 2023

[2] ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models 2019

[3] Are we done with imagenet? 2006

[4] Bet- ter plain vit baselines for imagenet-1k, 2022 2022

[5] Lucas Beyer, Xiaohua Zhai, and Alexander Kolesnikov. Big vision. https://github.com/google-research/ big_vision, 2022. 10, 17 2022

Cited by

31 papers in Pith

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

Is Dimensionality a Barrier for Retrieval Models?

A Gaia-linked High-purity QSO Candidate Catalog in Selected Fields with Extinction-binned Calibration and Spectrum-informed Training

FLARE: Fully Integration of Vision-Language Representations for Deep Cross-Modal Understanding

Policy Contrastive Decoding for Robotic Foundation Models

Receipt and verification

First computed	2026-05-17T23:38:47.785603Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

bf5bcf098924c3c987a41fa690345a2d4c3f37b27b81749215a8edb78e20ed8c

Aliases

arxiv: 2303.15343 · arxiv_version: 2303.15343v4 · doi: 10.48550/arxiv.2303.15343 · pith_short_12: X5N46CMJETB4 · pith_short_16: X5N46CMJETB4TB5E · pith_short_8: X5N46CMJ

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/X5N46CMJETB4TB5ED6TJANC2FV \
  | 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: bf5bcf098924c3c987a41fa690345a2d4c3f37b27b81749215a8edb78e20ed8c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "cc0332e47d7937841046a3090d20ffdd67b32b1eb76a47acb7de21d23a708af6",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2023-03-27T15:53:01Z",
    "title_canon_sha256": "7f4a056490916a9af113a28c02097641876e16d9ab2879231c7be248053b6bdb"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2303.15343",
    "kind": "arxiv",
    "version": 4
  }
}