Pith Number

pith:IL6PIF3P

pith:2026:IL6PIF3PTLE7CCQLVVVV25UXNN

not attested not anchored not stored refs resolved

Zero-Shot Faithful Textual Explanations via Directional-Derivative Influence on Predictions

Hiroshi Kera, Kazuhiko Kawamoto, Toshinori Yamauchi

FaithTrace generates more faithful zero-shot textual explanations for image classifiers by using directional derivatives of class logits in feature space as a faithfulness proxy.

arxiv:2605.16877 v1 · 2026-05-16 · cs.CV

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{IL6PIF3PTLE7CCQLVVVV25UXNN}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more

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

FaithTrace yields more faithful explanations than baselines by using an influence score computed as the directional derivative of the class logit along the text-induced direction in the classifier's feature space as a proxy for faithfulness.

C2weakest assumption

That the directional derivative of the class logit along a text-induced direction in feature space serves as a valid and sufficient proxy for the true faithfulness of the textual explanation to the model's actual decision process.

C3one line summary

FaithTrace uses the directional derivative of class logits along text-induced directions in feature space as an influence score to produce and evaluate more faithful zero-shot textual explanations for image classifiers.

References

48 extracted · 48 resolved · 2 Pith anchors

[1] 08774, 2023 1, 2 2023

[2] Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond 2023 · arXiv:2308.12966

[3] In: Advances in Neural Information Processing Sys- tems (NeurIPS) (2024) 2 2024

[4] In: 2017 IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR) 2017

[5] In: Proceedings of t he IEEE International Conference on Computer Vision (ICCV) (2021) 6 2021

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T00:03:27.782396Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

42fcf4176f9ac9f10a0bad6b5d76976b41df96ef2e21748cc56e33f68d8dc5c7

Aliases

arxiv: 2605.16877 · arxiv_version: 2605.16877v1 · doi: 10.48550/arxiv.2605.16877 · pith_short_12: IL6PIF3PTLE7 · pith_short_16: IL6PIF3PTLE7CCQL · pith_short_8: IL6PIF3P

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/IL6PIF3PTLE7CCQLVVVV25UXNN \
  | 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: 42fcf4176f9ac9f10a0bad6b5d76976b41df96ef2e21748cc56e33f68d8dc5c7

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "1485dd9a4d4f51ed8d8d1b4d223a325429cc0072dc0b6e8228cf65846756ce6e",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-16T08:39:05Z",
    "title_canon_sha256": "00bb8fa25e7bb4d3749e5a0e0a3d1521f4695ad245d5d055e8ed0622b41a62f3"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.16877",
    "kind": "arxiv",
    "version": 1
  }
}