Pith Number

pith:4TT7PDU2

pith:2026:4TT7PDU2ST3K2TTCID2DXH3T5H

not attested not anchored not stored refs pending

Linearizing Vision Transformer with Test-Time Training

Dongchen Han, Gao Huang, Hanyi Wang, Yining Li, Yulin Wang, Zeyu Liu

Test-Time Training aligns linear attention with pretrained Softmax weights, enabling transfer after minimal fine-tuning.

arxiv:2605.02772 v2 · 2026-05-04 · 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{4TT7PDU2ST3K2TTCID2DXH3T5H}

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

With only 1 hour of fine-tuning on 4×H20 GPUs, SD3.5-T^5 achieves comparable text-to-image quality to the fine-tuned Softmax model, while accelerating inference by 1.32× and 1.47× at 1K and 2K resolutions.

C2weakest assumption

The representational gap between Softmax and linear attention can be closed sufficiently by TTT's two-layer dynamic formulation plus the introduced key instance normalization and lightweight locality enhancement module to allow effective weight inheritance.

C3one line summary

Using Test-Time Training's structural match to Softmax attention plus key normalization and locality modules allows inheriting pretrained weights and fine-tuning Stable Diffusion 3.5 in one hour to match quality while speeding inference 1.32-1.47x.

Formal links

3 machine-checked theorem links

Receipt and verification

First computed	2026-05-29T01:05:11.529602Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

e4e7f78e9a94f6ad4e6240f43b9f73e9e0fba53aaa694895847f3ea0ef93094b

Aliases

arxiv: 2605.02772 · arxiv_version: 2605.02772v2 · doi: 10.48550/arxiv.2605.02772 · pith_short_12: 4TT7PDU2ST3K · pith_short_16: 4TT7PDU2ST3K2TTC · pith_short_8: 4TT7PDU2

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/4TT7PDU2ST3K2TTCID2DXH3T5H \
  | 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: e4e7f78e9a94f6ad4e6240f43b9f73e9e0fba53aaa694895847f3ea0ef93094b

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "c70e9ae07c03a95926855e51fc82a37964e7617dc3da0f0f41fa056adff106f7",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-04T16:16:26Z",
    "title_canon_sha256": "b71a814c742e400a2f004d83bceca10675a1a765736b26042028c90b8cdbaa39"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.02772",
    "kind": "arxiv",
    "version": 2
  }
}