Interpretability Techniques for Deep Learning

Benchmarks archive 2025-07-28

2 leaderboard tables shown for this task, 2 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted.

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. A dash means no Syntology record for that paper, not a recorded non-run. Read from the graph 2026-09-24.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

2 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

No subtask under this task in the archive's task tree.

Most implemented papers archive 2025-07-28

22 shown of 22 papers with code (25 tagged with this task in all), ordered by repositories listed in the archive, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers without a page here are shown as plain text.

• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
  7 Oct 2016 126 repositories listed Syntology ran 79 of 141 samples · 62 unverified · 68 pointer-only (licence)
  For captioning and VQA, we show that even non-attention based models can localize inputs.
• Axiomatic Attribution for Deep Networks
  4 Mar 2017 40 repositories listed Syntology ran 36 of 56 samples · 20 unverified · 17 pointer-only (licence)
  We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works.
• "Why Should I Trust You?": Explaining the Predictions of Any Classifier
  16 Feb 2016 27 repositories listed Syntology ran 5 of 19 samples · 14 unverified
  Despite widespread adoption, machine learning models remain mostly black boxes.
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
  20 Dec 2013 23 repositories listed Syntology ran 1 of 4 samples · 3 unverified · 1 pointer-only (licence)
  This paper addresses the visualisation of image classification models, learnt using deep Convolutional Networks (ConvNets).
• A Unified Approach to Interpreting Model Predictions
  22 May 2017 17 repositories listed Syntology ran 3 of 8 samples · 5 unverified · 6 pointer-only (licence)
  Understanding why a model makes a certain prediction can be as crucial as the prediction's accuracy in many applications.
• RISE: Randomized Input Sampling for Explanation of Black-box Models
  19 Jun 2018 13 repositories listed Syntology ran 1 of 34 samples · 33 unverified · 10 pointer-only (licence)
  We compare our approach to state-of-the-art importance extraction methods using both an automatic deletion/insertion metric and a pointing metric based on human-annotated object segments.
• Time series saliency maps: explaining models across multiple domains
  19 May 2025 2 repositories listed
  These results demonstrate the ability of cross-domain integrated gradients to provide semantically meaningful insights in time-series models that are impossible with traditional time-domain saliency.
• CausalGym: Benchmarking causal interpretability methods on linguistic tasks
  19 Feb 2024 2 repositories listed Syntology ran 2 of 2 samples · 0 unverified · 2 pointer-only (licence)
  Language models (LMs) have proven to be powerful tools for psycholinguistic research, but most prior work has focused on purely behavioural measures (e.
• TraceFL: Interpretability-Driven Debugging in Federated Learning via Neuron Provenance
  21 Dec 2023 2 repositories listed
  We introduce TraceFL, a fine-grained neuron provenance capturing mechanism that identifies clients responsible for a global model's prediction by tracking the flow of information from individual clients to the global…
• Exploration of Interpretability Techniques for Deep COVID-19 Classification using Chest X-ray Images
  3 Jun 2020 2 repositories listed
  The outbreak of COVID-19 has shocked the entire world with its fairly rapid spread and has challenged different sectors.
• What Do Compressed Deep Neural Networks Forget?
  13 Nov 2019 2 repositories listed
  However, this measure of performance conceals significant differences in how different classes and images are impacted by model compression techniques.
• Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretability
  26 Mar 2025 1 repository listed Syntology ran 1 of 1 samples · 0 unverified · 1 pointer-only (licence)
  We also dissect the discovered model mechanism, revealing different intrinsic features controlling fine-grained aspects of generation, boosting further research on mechanistic interpretability of diffusion models.
• IBO: Inpainting-Based Occlusion to Enhance Explainable Artificial Intelligence Evaluation in Histopathology
  29 Aug 2024 1 repository listed
  Our results demonstrate that IBO significantly improves perceptual fidelity, achieving nearly twice the improvement in LPIPS scores compared to the best existing occlusion strategy.
• Explainable Deep Learning: A Visual Analytics Approach with Transition Matrices
  29 Mar 2024 1 repository listed
  In this work, we propose a novel approach that utilizes a transition matrix to interpret results from DL models through more comprehensible machine learning (ML) models.
• Less is More: Fewer Interpretable Region via Submodular Subset Selection
  14 Feb 2024 1 repository listed Syntology ran 3 of 3 samples · 0 unverified · 3 pointer-only (licence)
  For incorrectly predicted samples, our method achieves gains of 81.
• Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure
  13 Jun 2022 1 repository listed Syntology ran 1 of 1 samples · 0 unverified · 1 pointer-only (licence)
  HSIC measures the dependence between regions of an input image and the output of a model based on kernel embeddings of distributions.
• Learning the Dynamics of Physical Systems from Sparse Observations with Finite Element Networks
  16 Mar 2022 1 repository listed Syntology ran 0 of 8 samples · 8 unverified
  We propose a new method for spatio-temporal forecasting on arbitrarily distributed points.
• A Novel Deep Learning Model for Hotel Demand and Revenue Prediction amid COVID-19
  8 Mar 2022 1 repository listed
  To this end, it is essential to develop an interpretable forecast model that supports managerial and organizational decision-making.
• A Semi-supervised Deep Transfer Learning Approach for Rolling-Element Bearing Remaining Useful Life Prediction
  29 Sep 2021 1 repository listed
  Deep learning techniques have recently brought many improvements in the field of neural network training, especially for prognosis and health management.
• DISSECT: Disentangled Simultaneous Explanations via Concept Traversals
  31 May 2021 1 repository listed Syntology ran 0 of 1 samples · 1 unverified · 1 pointer-only (licence)
  Explaining deep learning model inferences is a promising venue for scientific understanding, improving safety, uncovering hidden biases, evaluating fairness, and beyond, as argued by many scholars.
• DeepNNK: Explaining deep models and their generalization using polytope interpolation
  20 Jul 2020 1 repository listed
  Modern machine learning systems based on neural networks have shown great success in learning complex data patterns while being able to make good predictions on unseen data points.
• Contextual Explanation Networks
  29 May 2017 1 repository listed
  Our results on image and text classification and survival analysis tasks demonstrate that CENs are not only competitive with the state-of-the-art methods but also offer additional insights behind each prediction, that…

Syntology lines on 12 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced. Read from the graph 2026-09-24.