REVIEW 5 cited by
Explaining Black Box Predictions and Unveiling Data Artifacts through Influence Functions
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Modern deep learning models for NLP are notoriously opaque. This has motivated the development of methods for interpreting such models, e.g., via gradient-based saliency maps or the visualization of attention weights. Such approaches aim to provide explanations for a particular model prediction by highlighting important words in the corresponding input text. While this might be useful for tasks where decisions are explicitly influenced by individual tokens in the input, we suspect that such highlighting is not suitable for tasks where model decisions should be driven by more complex reasoning. In this work, we investigate the use of influence functions for NLP, providing an alternative approach to interpreting neural text classifiers. Influence functions explain the decisions of a model by identifying influential training examples. Despite the promise of this approach, influence functions have not yet been extensively evaluated in the context of NLP, a gap addressed by this work. We conduct a comparison between influence functions and common word-saliency methods on representative tasks. As suspected, we find that influence functions are particularly useful for natural language inference, a task in which 'saliency maps' may not have clear interpretation. Furthermore, we develop a new quantitative measure based on influence functions that can reveal artifacts in training data.
Forward citations
Cited by 5 Pith papers
-
Newfluence: Boosting Model interpretability and Understanding in High Dimensions
In high-dimensional regression, classical influence functions underestimate true leave-one-out influence by a per-point factor, and the proposed Newfluence estimator corrects this bias.
-
Fair Document Valuation in LLM Summaries via Shapley Values
Cluster Shapley groups semantically similar documents via embeddings and computes cluster-level Shapley values, claiming better efficiency-accuracy trade-offs than Monte Carlo and Kernel SHAP on Amazon review summarization.
-
A Comparative Analysis of Influence Signals for Data Debugging
A benchmark shows Self-Influence finds mislabeled training samples, while all tested influence signals fail to detect clustered anomalies and outliers under cumulative TracIn scoring.
-
destroR: A Benchmark and Adversarial-Training Defense for Bangla Transfer Models under Meaning-Preserving Attacks
Claims a Bangla adversarial-attack benchmark and defense, but the text delivers a single-model attack study with no defense and no baselines.
-
Attributing Data for Sharpness-Aware Minimization
SAM-HIF and SAM-GIF are proposed as data attribution scores for SAM-trained models, but SAM-GIF is TracIn with SAM gradients and SAM-HIF's derivation contains a load-bearing error.
Discussion (0). Sign in to comment.