Pith Number

pith:7RAZXCUC

pith:2026:7RAZXCUCFI2IW76ZLHTPBBCK63

not attested not anchored not stored refs resolved

DIVER:Diving Deeper into Distilled Data via Expressive Semantic Recovery

Guoming Lu, Jiawei Du, Jielei Wang, Qianxin Xia, Wenbo Jiang, Zhiyong Shu

A dual-stage framework uses a pre-trained diffusion model to recover expressive semantics from distilled datasets and improve performance across different neural architectures.

arxiv:2605.12649 v1 · 2026-05-12 · 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{7RAZXCUCFI2IW76ZLHTPBBCK63}

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

DIVER leverages the pre-trained diffusion model to dive deeper into distilled data via expressive semantic recovery, an entire process of semantic inheritance, guidance, and fusion... significantly improving cross-architecture generalization, requiring processing time comparable to raw DiT on ImageNet (256×256) with only 4 GB of GPU memory usage.

C2weakest assumption

That the pre-trained diffusion model can reliably filter architecture-specific noise in the latent space while preserving intrinsic semantics, and that applying semantic guidance only in the concrete phase of the reverse process avoids ambiguity and artifacts without losing essential information.

C3one line summary

DIVER is a dual-stage distillation method using diffusion models to enhance semantic preservation and cross-architecture generalization in dataset distillation.

References

62 extracted · 62 resolved · 5 Pith anchors

[1] Distilling the Knowledge in a Neural Network · arXiv:1503.02531

[2] Dataset distillation

[3] Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages=

[4] Dataset condensation with gradient matching 2006

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

Receipt and verification

First computed	2026-05-18T03:09:59.770792Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

fc419b8a822a348b7fd959e6f0844af6d4a1493491505f53ed5c50ac3d1f32cb

Aliases

arxiv: 2605.12649 · arxiv_version: 2605.12649v1 · doi: 10.48550/arxiv.2605.12649 · pith_short_12: 7RAZXCUCFI2I · pith_short_16: 7RAZXCUCFI2IW76Z · pith_short_8: 7RAZXCUC

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/7RAZXCUCFI2IW76ZLHTPBBCK63 \
  | 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: fc419b8a822a348b7fd959e6f0844af6d4a1493491505f53ed5c50ac3d1f32cb

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "346c8f4b8fbe2d7d2891e76533ae1003f0e67aa7a496d8b9bccc35814bae744e",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-12T18:55:53Z",
    "title_canon_sha256": "3bd5c90b553bf9c6aca7aecc870c7fdcd628968e7882aff88046ac1863e49da6"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12649",
    "kind": "arxiv",
    "version": 1
  }
}