Papers › Axiomatic Attribution for Deep Networks

Axiomatic Attribution for Deep Networks

4 Mar 2017ICML 2017 8arXiv:1703.01365archive 2025-07-28

Mukund Sundararajan, Ankur Taly, Qiqi Yan

We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works. We identify two fundamental axioms---Sensitivity and Implementation Invariance that attribution methods ought to satisfy. We show that they are not satisfied by most known attribution methods, which we consider to be a fundamental weakness of those methods. We use the axioms to guide the design of a new attribution method called Integrated Gradients. Our method requires no modification to the original network and is extremely simple to implement; it just needs a few calls to the standard gradient operator. We apply this method to a couple of image models, a couple of text models and a chemistry model, demonstrating its ability to debug networks, to extract rules from a network, and to enable users to engage with models better.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 56 code samples harvested from 21 repositories linked to this paper: 36 ran (of which 13 constructed an object rather than computing a result; 21 with no instrument failure: 3 honoured, 1 violated, 17 with no contract checked; 15 where Syntology's instrument failed) · 20 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 8 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 55 samples from 20 repositories, 35 ran (21 with no instrument failure; 14 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 8 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

40 repositories listed; official and paper-mentioned ones first.

ankurtaly/Attributions officialmentioned in papermentioned on GitHubtfSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
AlejandroAttento/Pytorch-Captum mentioned on GitHubpytorchSyntology: not harvested report
JoHof/IntegratedGradientsTutorial mentioned on GitHubSyntology: not harvested report
TianhongDai/integrated-gradient-pytorch mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
TooTouch/WhiteBox-Part1 mentioned on GitHubpytorchSyntology: 3 samples, 2 ran (1 instrument-failed) · 1 unverified report
allenwind/text-integrated-gradients mentioned on GitHubtfSyntology: not harvested report
andresbecker/master_thesis mentioned on GitHubtfSyntology: 4 samples, 3 ran (1 instrument-failed) · 1 unverified report
ascillitoe/shap mentioned on GitHubtfSyntology: not harvested report
austinbrown34/shap mentioned on GitHubtfSyntology: 8 samples, 6 ran (3 instrument-failed) · 2 unverified report
bips-hb/innsight mentioned on GitHubtorchSyntology: not harvested report
cdpierse/transformers-interpret mentioned on GitHubpytorchSyntology: 3 samples, 1 ran (0 instrument-failed) · 2 unverified report
fcUalberta/UAlberta-Multimedia-Masters-Program-Interpretable-AI-Part_1_2 mentioned on GitHubtfSyntology: 1 sample, 0 ran · 1 unverified report
gablabc/shap mentioned on GitHubtfSyntology: not harvested report
galdeia/iirsbenchmark mentioned on GitHubSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
garygsw/smooth-taylor mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
hannamw/eap-ig mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models (arXiv:2403.19647; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified) and Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attribution (arXiv:2605.15138; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
jemilc/shap mentioned on GitHubtfSyntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
koren-v/Interpret mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
marnifora/magisterka mentioned on GitHubpytorchSyntology: 3 samples, 1 ran (1 instrument-failed) · 2 unverified report
miaolan-xie/shap mentioned on GitHubtfSyntology: not harvested report
nsaphra/acd mentioned on GitHubpytorchSyntology: 5 samples, 4 ran (2 instrument-failed) · 1 unverified report
pamflecista/Magisterka mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
pytorch/captum mentioned on GitHubpytorchBSD-3-ClauseSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Understanding and Improving Feature Learning for Out-of-Distribution Generalization (arXiv:2304.11327; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
saivarunr/xshap mentioned on GitHubtfSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
shaoshanglqy/shap-shapley mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Learning Important Features Through Propagating Activation Differences (arXiv:1704.02685; that paper's own run record: 8 ran (of which 2 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 1 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified) and Consistent Individualized Feature Attribution for Tree Ensembles (arXiv:1802.03888; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
shap/shap mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Hide&Seek: Learning to Explain in an End-to-End Differentiable Network (arXiv:2608.16689; that paper's own run record: 12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 5 honoured, 1 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
shyhyawJou/Integrated-Gradient-Pytorch mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
shyhyawJou/Integrated-Gradient-Tensorflow mentioned on GitHubtfSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
sicara/tf-explain mentioned on GitHubtfSyntology: 4 samples, 2 ran (2 instrument-failed) · 2 unverified report
suinleelab/path_explain mentioned on GitHubtfSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
tleemann/road_evaluation mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to A Consistent and Efficient Evaluation Strategy for Attribution Methods (arXiv:2202.00449; that paper's own run record: 10 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
tomdyer10/fake_news mentioned on GitHubpytorchSyntology: not harvested report
uhussai7/boldreams mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
yeefan1999/Explainable-Health-Prediction-with-Transfer-Learning mentioned on GitHubtfSyntology: not harvested report
MindSpore-scientific-2/code-4 mindsporeSyntology: not harvested report
MindSpore-scientific/code-8 mindsporeApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SimBa: Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning (arXiv:2410.09754; that paper's own run record: 6 ran (of which 2 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
jankrepl/mildlyoverfitted jaxSyntology: 3 samples, 3 ran (3 instrument-failed) · 0 unverified report
mindspore-ai/contrib mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems (arXiv:2305.12102; that paper's own run record: 2 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 1 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
pwc-1/Paper-9 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RandAugment: Practical automated data augmentation with a reduced search space (arXiv:1909.13719; that paper's own run record: 58 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 51 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

56 samples harvested: 36 ran (of which 13 constructed an object rather than computing a result; 21 with no instrument failure: 3 honoured, 1 violated, 17 with no contract checked; 15 where Syntology's instrument failed) · 20 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
1ran · violated contract
7ran · our draft was wrong
8ran; the contract check could not run
17ran
20unverified

Licence: 17 of the 56 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 21 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

ankurtaly/Attributions official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
integrated_gradients in 5 papers IntegratedGradients/integrated_gradients.py ran · our draft was wrong no licence file found · pointer only · f36f0780dc332c6d · report fixture: signature inp:ndarray/2/float, target_label_index:int/0/int, predictions_and_gradients:callable/0/?, baseline:ndarray/2/float, steps:int/0/int · returned tuple · level recorded 2026-09-22
TianhongDai/integrated-gradient-pytorch community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
pre_processing in 5 papers integrated_gradients.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · f4fdca61674e575a · report fixture: signature obs:ndarray/3/float, cuda:bool/0/bool · returned Tensor · level recorded 2026-09-22
integrated_gradients in 5 papers integrated_gradients.py unverified MIT (permissive) · 04f60f57d1b6028e · report
TooTouch/WhiteBox-Part1 community (archive-listed) 3 samples, 2 ran (1 instrument-failed) · 1 unverified
IntegratedGradients code/saliency/attribution_methods.py ran MIT (permissive) · 67d9ff48d5643c84 · report fixture: returned IntegratedGradients · level recorded 2026-09-22
rescale_image code/saliency/attribution_methods.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · e44a9afe5dc5a201 · report fixture: signature images:ndarray/4/float, channel:bool/0/bool · returned ndarray · level recorded 2026-09-22
save_saliency_map code/saliency/attribution_methods.py unverified MIT (permissive) · c91220cf9e079883 · report
andresbecker/master_thesis community (archive-listed) 4 samples, 3 ran (1 instrument-failed) · 1 unverified
compute_gradients in 9 papers workspace/libs/NN_interpretability.py ran; the contract check could not run (RAISES) MIT (permissive) · 911004b0e426d584 · report fixture: signature images:ResourceVariable/2/float, model:callable/0/? · returned EagerTensor · level recorded 2026-09-22
get_integral_approximation workspace/libs/NN_interpretability.py ran · honoured contract fingerprinted MIT (permissive) · 76f2f497c06af59e · report fixture: signature gradients:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
interpolate_images workspace/libs/NN_interpretability.py ran · violated contract MIT (permissive) · de4b53a5e89b25d2 · report fixture: signature baseline:EagerTensor/3/float, image:EagerTensor/3/float, alphas:EagerTensor/1/float · returned EagerTensor · level recorded 2026-09-22
get_integrated_gradients workspace/libs/NN_interpretability.py unverified MIT (permissive) · 745ae6a587a2034e · report
austinbrown34/shap community (archive-listed) 8 samples, 6 ran (3 instrument-failed) · 2 unverified
GradientExplainer shap/explainers/gradient.py ran MIT (permissive) · e8960b01109d72e4 · report fixture: returned GradientExplainer · level recorded 2026-09-22
_PyTorchGradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 7374a66bd3df8287 · report fixture: returned _PyTorchGradientExplainer · level recorded 2026-09-22
_TFGradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 5031eabe21a685be · report fixture: returned _TFGradientExplainer · level recorded 2026-09-22
_get_model_inputs shap/explainers/gradient.py ran · our draft was wrong MIT (permissive) · 35d945c6546483b0 · report fixture: signature model:callable/0/? · returned list · level recorded 2026-09-22
_get_model_output shap/explainers/gradient.py ran · our draft was wrong MIT (permissive) · 6b8b878c41b5aa3c · report fixture: signature model:callable/0/? · returned KerasTensor · level recorded 2026-09-22
_get_session shap/explainers/gradient.py ran · our draft was wrong MIT (permissive) · f242e8dd0a93f71b · report fixture: signature session:Session/0/? · returned Session · level recorded 2026-09-22
_get_graph shap/explainers/gradient.py unverified MIT (permissive) · c9c180d15ae62447 · report
_import_tf shap/explainers/gradient.py unverified MIT (permissive) · 8fc8b96ad1c4dff5 · report
cdpierse/transformers-interpret community (archive-listed) 3 samples, 1 ran (0 instrument-failed) · 2 unverified
Attributions transformers_interpret/attributions.py ran Apache-2.0 (permissive) · ef35ae3770394c9b · report fixture: returned Attributions · level recorded 2026-09-22
AttributionsNotCalculatedError transformers_interpret/attributions.py unverified Apache-2.0 (permissive) · 8900360e60786d71 · report
LIGAttributions transformers_interpret/attributions.py unverified Apache-2.0 (permissive) · ebfc7e44821a7446 · report
fcUalberta/UAlberta-Multimedia-Masters-Program-Interpretable-AI-Part_1_2 community (archive-listed) 1 sample, 0 ran · 1 unverified
integrated_gradients in 5 papers code/IntegratedGradients.py unverified AGPL-3.0 (copyleft) · pointer only · b1ea75b6852183e8 · report
galdeia/iirsbenchmark community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
Base_explainer iirsBenchmark/explainers/IntegratedGradients_explainer.py ran BSD-3-Clause (permissive) · ecdb335eb9bb79a3 · report fixture: returned Base_explainer · level recorded 2026-09-22
IntegratedGradients_explainer iirsBenchmark/explainers/IntegratedGradients_explainer.py ran BSD-3-Clause (permissive) · 8e17779e3b980afa · report fixture: returned IntegratedGradients_explainer · level recorded 2026-09-22
NotApplicableException iirsBenchmark/explainers/IntegratedGradients_explainer.py ran BSD-3-Clause (permissive) · 06c3310a0f107606 · report fixture: returned NotApplicableException · level recorded 2026-09-22
garygsw/smooth-taylor community (archive-listed) 2 samples, 0 ran · 2 unverified
get_gradients attribution/analyzer.py unverified MIT (permissive) · 942fe31074420d3b · report
integrated_gradients in 5 papers attribution/analyzer.py unverified MIT (permissive) · 825842184e64aa86 · report
jemilc/shap community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
GradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 898260b979b40803 · report fixture: returned GradientExplainer · level recorded 2026-09-22
_PyTorchGradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 430707dd56b72369 · report fixture: returned _PyTorchGradientExplainer · level recorded 2026-09-22
_TFGradientExplainer shap/explainers/gradient.py unverified MIT (permissive) · 9326aa10a22b3257 · report
koren-v/Interpret community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
IntegratedGradient integrated_gradient.py ran no licence file found · pointer only · 57137c7c74c0ba8b · report fixture: returned IntegratedGradient · level recorded 2026-09-22
SaliencyInterpreter integrated_gradient.py ran no licence file found · pointer only · 7e1b42c1361193d8 · report fixture: returned SaliencyInterpreter · level recorded 2026-09-22
marnifora/magisterka community (archive-listed) 3 samples, 1 ran (1 instrument-failed) · 2 unverified
calculate_gradients bin/integrated_gradients.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed no licence file found · pointer only · 7f1cc66f1073bcfd · report fixture: signature model:callable/0/?, inputs:list/1/Tensor, labels:list/1/Tensor, use_cuda:bool/0/bool · returned ndarray · level recorded 2026-09-22
OHEncoder bin/integrated_gradients.py unverified no licence file found · pointer only · 4ddae2915b9cc97f · report
integrated_gradients in 5 papers bin/integrated_gradients.py unverified no licence file found · pointer only · 6f457ae7012a204f · report
nsaphra/acd community (archive-listed) 5 samples, 4 ran (2 instrument-failed) · 1 unverified
calc_pad acd/scores/score_funcs.py ran · honoured contract fingerprinted no licence file found · pointer only · 54e4782533ba8866 · report fixture: signature pad:str/0/str, in_siz:int/0/int, out_siz:int/0/int, stride:int/0/int, ksize:int/0/int · returned int · level recorded 2026-09-22
calc_size acd/scores/score_funcs.py ran · honoured contract no licence file found · pointer only · 4100b072ce597ce6 · report fixture: signature h:int/0/int, kh:int/0/int, pad:str/0/str, sh:int/0/int · returned float64 · level recorded 2026-09-22
conv2dnp acd/scores/score_funcs.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant no licence file found · pointer only · eb27e71fedd48267 · report fixture: signature x:ndarray/4/float, w:ndarray/4/float, pad:str/0/str, stride:tuple/1/int · returned ndarray · level recorded 2026-09-22
extract_sliding_windows acd/scores/score_funcs.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant no licence file found · pointer only · e9b75455324ec1d2 · report fixture: signature x:ndarray/4/float, ksize:list/1/int, pad:str/0/str, stride:list/1/int, floor_first:bool/0/bool · returned ndarray · level recorded 2026-09-22
ig_scores_2d acd/scores/score_funcs.py unverified no licence file found · pointer only · d615bfc8c34fd228 · report
pamflecista/Magisterka community (archive-listed) 2 samples, 0 ran · 2 unverified
OHEncoder bin/integrated_gradients.py unverified no licence file found · pointer only · c58da49866be23b4 · report
integrated_gradients in 5 papers bin/integrated_gradients.py unverified no licence file found · pointer only · c69ac1fe40cb16f0 · report
saivarunr/xshap community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
GradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 6cc4d6985789886b · report fixture: signature X:Tensor/2/float, nsamples:int/0/int, ranked_outputs:NoneType/0/?, output_rank_order:str/0/str · returned list · level recorded 2026-09-24
_PyTorchGradientExplainer shap/explainers/gradient.py ran MIT (permissive) · 2afe9377139c8f9a · report fixture: signature X:Tensor/2/float, nsamples:int/0/int, ranked_outputs:NoneType/0/?, output_rank_order:str/0/str · returned list · level recorded 2026-09-24
shyhyawJou/Integrated-Gradient-Pytorch community (archive-listed) 1 sample, 0 ran · 1 unverified
Integrated_Gradient ig.py unverified GPL-3.0 (copyleft) · pointer only · f266368a15af5029 · report
shyhyawJou/Integrated-Gradient-Tensorflow community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
Integrated_Gradient ig.py ran GPL-3.0 (copyleft) · pointer only · 66d0f363e78a65b7 · report fixture: signature img:ndarray/3/int, baseline:NoneType/0/? · returned tuple · level recorded 2026-09-24
sicara/tf-explain community (archive-listed) 4 samples, 2 ran (2 instrument-failed) · 2 unverified
grid_display tf_explain/core/integrated_gradients.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 2fd99b461cb809d0 · report fixture: signature array:ndarray/4/int, num_rows:NoneType/0/?, num_columns:NoneType/0/? · returned ndarray · level recorded 2026-09-22
transform_to_normalized_grayscale tf_explain/core/integrated_gradients.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · bba688b42ae9d0fc · report fixture: signature tensor:EagerTensor/4/float · returned EagerTensor · level recorded 2026-09-22
IntegratedGradients tf_explain/core/integrated_gradients.py unverified MIT (permissive) · 3785a3b3dee4de2b · report
save_grayscale tf_explain/core/integrated_gradients.py unverified MIT (permissive) · 73fe690c71ad066c · report
suinleelab/path_explain community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
PathExplainerTF path_explain/explainers/path_explainer_tf.py ran MIT (permissive) · 4c4100f237fa3e0c · report fixture: returned PathExplainerTF · level recorded 2026-09-22
Explainer in 9 papers path_explain/explainers/path_explainer_tf.py unverified MIT (permissive) · 543fdd3d86e7ad28 · report
uhussai7/boldreams community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
integrated_gradient models/modules.py ran no licence file found · pointer only · 41eeb813009c9224 · report fixture: signature img_in:Tensor/4/float, roi:int/0/int, I_0:NoneType/0/?, steps:int/0/int · returned Tensor · level recorded 2026-09-24
jankrepl/mildlyoverfitted community (archive-listed) 3 samples, 3 ran (3 instrument-failed) · 0 unverified
compute_gradient in 10 papers mini_tutorials/gradient_wrt_input/explain.py ran · our draft was wrong MIT (permissive) · 0ed09db5a4a023e1 · report fixture: signature func:callable/0/?, inp:Tensor/2/float, net:callable/0/?, target:int/0/int · returned Tensor · level recorded 2026-09-22
compute_integrated_gradients mini_tutorials/gradient_wrt_input/explain.py ran · our draft was wrong MIT (permissive) · df531c40d05d60e2 · report fixture: signature inp:Tensor/4/float, baseline:Tensor/4/float, net:callable/0/?, target:int/0/int, n_steps:int/0/int · returned tuple · level recorded 2026-09-22
func in 29 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 4 buckets, one shared input per bucket
mini_tutorials/gradient_wrt_input/explain.py ran · our draft was wrong MIT (permissive) · 671a764793394b27 · report fixture: signature inp:Tensor/2/float, net:callable/0/?, target:int/0/int · returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

21 assemblies on this paper: 21 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

21 sibling assemblies for Integrated Gradients (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another. Bricks printed for 12 of 21 sibling assemblies; the other 9 are listed with their counts; every brick is in this page's JSON twin.

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 3 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. integrated_gradients entry · level 0 · unverified · 6f457ae7012a204f
  2. OHEncoder data_transform · level 0 · unverified · 4ddae2915b9cc97f
  3. calculate_gradients helper · level 1 · ran; the contract check could not run (RAISES) · 7f1cc66f1073bcfd

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. integrated_gradient entry · level 1 · ran · 41eeb813009c9224

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 3 bricks; individually on a fixture: 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. IntegratedGradients entry · level 1 · ran · 67d9ff48d5643c84
  2. rescale_image helper · level 1 · ran; the contract check could not run (RAISES) · e44a9afe5dc5a201
  3. save_saliency_map helper · level 0 · unverified · c91220cf9e079883

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 5 bricks; individually on a fixture: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. ig_scores_2d entry · level 0 · unverified · d615bfc8c34fd228
  2. conv2dnp helper · level 1 · ran; the contract check could not run (RAISES) · eb27e71fedd48267
  3. extract_sliding_windows helper · level 1 · ran; the contract check could not run (RAISES) · e9b75455324ec1d2
  4. calc_pad helper · level 1 · ran · honoured contract · 54e4782533ba8866
  5. calc_size helper · level 1 · ran · honoured contract · 4100b072ce597ce6

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 2 bricks; individually on a fixture: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 2 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 21 siblings

  1. IntegratedGradient entry · level 1 · ran · 57137c7c74c0ba8b
  2. SaliencyInterpreter helper · level 1 · ran · 7e1b42c1361193d8

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 2 bricks; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 21 siblings

  1. PathExplainerTF entry · level 1 · ran · 4c4100f237fa3e0c
  2. Explainer helper · level 0 · unverified · 543fdd3d86e7ad28

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 4 bricks; individually on a fixture: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 2 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 21 siblings

  1. GradientExplainer entry · level 1 · ran · 898260b979b40803
  2. Explainer helper · level 0 · unverified · 6cddc3a37d231206
  3. _PyTorchGradientExplainer helper · level 1 · ran · 430707dd56b72369
  4. _TFGradientExplainer helper · level 0 · unverified · 9326aa10a22b3257

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 3 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. compute_integrated_gradients entry · level 1 · ran · our draft was wrong · df531c40d05d60e2
  2. compute_gradient helper · level 1 · ran · our draft was wrong · 0ed09db5a4a023e1
  3. func model · level 1 · ran · our draft was wrong · 671a764793394b27

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 4 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. get_integrated_gradients entry · level 0 · unverified · 745ae6a587a2034e
  2. compute_gradients helper · level 1 · ran; the contract check could not run (RAISES) · 911004b0e426d584
  3. get_integral_approximation helper · level 1 · ran · honoured contract · 76f2f497c06af59e
  4. interpolate_images helper · level 1 · ran · violated contract · de4b53a5e89b25d2

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 9 bricks; individually on a fixture: 6 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. GradientExplainer entry · level 1 · ran · e8960b01109d72e4
  2. Explainer helper · level 0 · unverified · 6cddc3a37d231206
  3. _PyTorchGradientExplainer helper · level 1 · ran · 7374a66bd3df8287
  4. _TFGradientExplainer helper · level 1 · ran · 5031eabe21a685be
  5. _get_graph helper · level 0 · unverified · c9c180d15ae62447
  6. _get_model_inputs helper · level 1 · ran · our draft was wrong · 35d945c6546483b0
  7. _get_model_output helper · level 1 · ran · our draft was wrong · 6b8b878c41b5aa3c
  8. _get_session helper · level 1 · ran · our draft was wrong · f242e8dd0a93f71b
  9. _import_tf helper · level 0 · unverified · 8fc8b96ad1c4dff5

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 21 siblings

  1. integrated_gradients entry · level 0 · unverified · b1ea75b6852183e8

harvested Assembly for Integrated Gradients (graph method: Integrated Gradients). 3 bricks; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 21 siblings

  1. LIGAttributions entry · level 0 · unverified · ebfc7e44821a7446
  2. Attributions helper · level 1 · ran · ef35ae3770394c9b
  3. AttributionsNotCalculatedError helper · level 0 · unverified · 8900360e60786d71
  1. integrated_gradients · 7502058c139f2ca3 · 2 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified · The assembly has not been run end-to-end · assemblies[12] in the JSON twin
  2. integrated_gradients · 94638f8263a79ecf · 2 bricks; individually on a fixture: 0 ran · 2 unverified · The assembly has not been run end-to-end · assemblies[13] in the JSON twin
  3. IntegratedGradients · 992b51655197bbc3 · 4 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified · The assembly has not been run end-to-end · assemblies[14] in the JSON twin
  4. IntegratedGradients_explainer · b6ca4edb5060f25a · 3 bricks; individually on a fixture: 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 3 samples that ran constructed an object rather than computing a result · The assembly has not been run end-to-end · assemblies[15] in the JSON twin
  5. Integrated_Gradient · be12adb6126e9dd2 · 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified · The assembly has not been run end-to-end · assemblies[16] in the JSON twin
  6. integrated_gradients · d89ace143fc7a244 · 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified · The assembly has not been run end-to-end · assemblies[17] in the JSON twin
  7. integrated_gradients · ec73ff54e5a512ec · 3 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified · The assembly has not been run end-to-end · assemblies[18] in the JSON twin
  8. GradientExplainer · f558b9347a05d10a · 4 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified · The assembly has not been run end-to-end · assemblies[19] in the JSON twin
  9. Integrated_Gradient · f73712f050775910 · 1 brick; individually on a fixture: 0 ran · 1 unverified · The assembly has not been run end-to-end · assemblies[20] in the JSON twin

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="1703.01365") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Explainable artificial intelligenceImage AttributionInterpretability Techniques for Deep LearningInterpretable Machine Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Attribution CUB-200-2011 Integrated Gradients Deletion AUC score (ResNet-101) 0.0728 #8 of 8 Archive leaderboard report
Image Attribution CUB-200-2011 Integrated Gradients Insertion AUC score (ResNet-101) 0.0422 #8 of 8 Archive leaderboard report
Image Attribution CelebA Integrated Gradients Deletion AUC score (ArcFace ResNet-101) 0.0680 #8 of 8 Archive leaderboard report
Image Attribution CelebA Integrated Gradients Insertion AUC score (ArcFace ResNet-101) 0.3578 #8 of 8 Archive leaderboard report
Image Attribution VGGFace2 Integrated Gradients Deletion AUC score (ArcFace ResNet-101) 0.0749 #7 of 8 Archive leaderboard report
Image Attribution VGGFace2 Integrated Gradients Insertion AUC score (ArcFace ResNet-101) 0.5399 #7 of 8 Archive leaderboard report
Interpretability Techniques for Deep Learning CelebA Integrated Gradients Insertion AUC score 0.3578 #7 of 7 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections