Pith Number

pith:PI7DO5NW

pith:2025:PI7DO5NWRCMOEM745WDBN73K7Y

not attested not anchored not stored refs pending

From Per-Image Low-Rank to Encoding Mismatch: Rethinking Feature Distillation in Vision Transformers

Bonan Xu, Huiyuan Tian, Shijian Li

An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers.

arxiv:2511.15572 v3 · 2025-11-19 · 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{PI7DO5NWRCMOEM745WDBN73K7Y}

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 refer to this combined phenomenon as an encoding mismatch. We propose two minimal remedies, Lift or WideLast... On ImageNet-1K, these fixes revive feature KD for ViT compression, improving DeiT-Tiny distilled from CaiT-S24 from 74.86% to 77.53%/78.23% top-1 accuracy.

C2weakest assumption

The assumption that the observed per-image low-rank structure, dataset-level subspace rotations, and token-level spectral energy patterns are the primary and causal reasons for feature-map KD failure in compression, rather than other factors such as optimization dynamics or capacity limits.

C3one line summary

The paper identifies an encoding mismatch in ViT feature distillation from per-image compressibility versus dataset subspace rotations and broad spectral energy patterns, proposing Lift and WideLast remedies that improve DeiT-Tiny accuracy from 74.86% to 77.53-78.23% on ImageNet-1K.

Cited by

1 paper in Pith

Attention Transfer Is Not Universally Effective for Vision Transformers

Receipt and verification

First computed	2026-05-27T02:05:08.393777Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc

Aliases

arxiv: 2511.15572 · arxiv_version: 2511.15572v3 · doi: 10.48550/arxiv.2511.15572 · pith_short_12: PI7DO5NWRCMO · pith_short_16: PI7DO5NWRCMOEM74 · pith_short_8: PI7DO5NW

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/PI7DO5NWRCMOEM745WDBN73K7Y \
  | 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: 7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "119b4c5982a6a6abd874ce894e60e69d93cad4c8d187812e7ea4e5ba9e01feea",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2025-11-19T16:03:21Z",
    "title_canon_sha256": "f1c33b217f031c57c0dbc57daea71266dd59ce186ae3ee4396e325d8ce3cd7cb"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2511.15572",
    "kind": "arxiv",
    "version": 3
  }
}