Pith Number

pith:G7PPI4VK

pith:2026:G7PPI4VK5CLXCTG454CHOAT3X3

not attested not anchored not stored refs resolved

AGOP as Explanation: From Feature Learning to Per-Sample Attribution in Image Classifiers

Raj Kiran Gupta Katakam

The Average Gradient Outer Product matrix from training data supplies a prior that improves per-sample attribution maps in image classifiers.

arxiv:2605.12816 v1 · 2026-05-12 · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{G7PPI4VK5CLXCTG454CHOAT3X3}

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

AGOP-Weighted achieves 44% higher mIoU than IG on linear tasks; AGOP-Global achieves 7x higher mIoU than IG on multiplicative tasks (where IG falls below random) at zero inference cost. Both findings generalise to ResNet-18 on CLEVR-XAI (+18% and +37% respectively).

C2weakest assumption

That the AGOP matrix computed over the training distribution supplies an unbiased prior that reliably suppresses gradient noise for individual test samples without introducing systematic errors when the test distribution differs from training.

C3one line summary

AGOP-based attribution methods outperform Integrated Gradients and other baselines on pixel-level ground truth benchmarks for explaining image classifier decisions, with AGOP-Global offering zero inference cost.

References

15 extracted · 15 resolved · 2 Pith anchors

[1] M. Sundararajan, A. Taly, Q. Yan, Axiomatic attribution for deep networks, in: Proceedings of ICML 2017, 2017, pp. 3319–3328 2017

[2] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, D. Batra, Grad-CAM: Visual explanations from deep networks via gradient-based localization, in: Proceedings of ICCV 2017, 2017, pp. 618–62 2017

[3] SmoothGrad: removing noise by adding noise 2017 · arXiv:1706.03825

[4] A. Radhakrishnan, D. Beaglehole, P. Pandit, M. Belkin, Mechanism for feature learning in neural networks and backpropagation-free machine learning models, Science 383 (2024) 1461–1467 2024

[5] D. Beaglehole, A. Radhakrishnan, P. Pandit, M. Belkin, Mechanism of feature learning in convolu- tional neural networks, arXiv preprint arXiv:2309.00570 (2024) 2024

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

AGOP-IxG: A Gradient Covariance Filter for Local Feature Attribution on Tabular Data, with a Controlled Benchmark

Receipt and verification

First computed	2026-05-18T03:09:12.302212Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

37def472aae897714cdcef0477027bbec019eac802dd4fb9ae4b763f9203ee2c

Aliases

arxiv: 2605.12816 · arxiv_version: 2605.12816v1 · doi: 10.48550/arxiv.2605.12816 · pith_short_12: G7PPI4VK5CLX · pith_short_16: G7PPI4VK5CLXCTG4 · pith_short_8: G7PPI4VK

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/G7PPI4VK5CLXCTG454CHOAT3X3 \
  | 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: 37def472aae897714cdcef0477027bbec019eac802dd4fb9ae4b763f9203ee2c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "fb65d6c22827edb69f9f9fc02a3970a1698dfd8a729b0ab73abc63a58d52d6b3",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-12T23:15:47Z",
    "title_canon_sha256": "2a6eff238dddf7c3b1b3cb6187fe01503e90a704b0dd362010a2622678e6b789"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12816",
    "kind": "arxiv",
    "version": 1
  }
}