| Challenge: | Modern deep learning models are notoriously opaque, which has motivated the development of methods for interpreting how deep models predict. |
| Approach: | They propose to review existing methods for evaluating attribution scores and summarize the logic traps in these methods. |
| Outcome: | The proposed methods show that they do not contain logic traps and that they are not reliable. |
Similar Papers
Explaining Pre-Trained Language Models with Attribution Scores: An Analysis in Low-Resource Settings (2024.lrec-main)
Copied to clipboard
| Challenge: | Currently, prompt-based models are gaining popularity due to their easier adaptability in low-resource settings. |
| Approach: | They analyze attribution scores extracted from prompt-based models w.r.t. plausibility and faithfulness and compare them with attribution score extracted from fine-tuned models and large language models. |
| Outcome: | The proposed model outperforms attention and Integrated Gradients in plausibility and faithfulness, while fine-tuning models are harder to explain in low-resource settings. |
“Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing work on faithfulness evaluation is not conclusive and does not provide a clear answer as to how different methods are to be compared. |
| Approach: | They propose a protocol for faithfulness evaluation that makes use of partially synthetic data to obtain ground truth for feature importance ranking. |
| Outcome: | The proposed method is based on partially synthetic data and is compared with lexical shortcuts on a range of datasets and LSTM models. |
AttributionBench: How Hard is Automatic Attribution Evaluation? (2024.findings-acl)
Copied to clipboard
| Challenge: | generative search engines enhance the reliability of large language model responses by providing cited evidence. |
| Approach: | They propose to use a benchmark to evaluate whether a large language model supports the generated responses or not . |
| Outcome: | The proposed benchmark shows that even a fine-tuned GPT-3.5 only achieves around 80% macro-F1 under a binary classification formulation. |
Did the Model Understand the Question? (P18-1)
Copied to clipboard
| Challenge: | Using the notion of “attribution,” deep learning models often ignore important question terms. |
| Approach: | They propose techniques to analyze the sensitivity of a deep learning model to question words . they use attribution to generate adversarial questions using visual and tabular questions . |
| Outcome: | The proposed techniques reduce the accuracy of a visual question answering model by 61.1% and that of 'tabular' question answering models by 3.3%. |
Diagnosing Failures in Large Language Models’ Answers: Integrating Error Attribution into Evaluation Framework (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation models lack error attribution capability due to their proprietary nature. |
| Approach: | They propose a misattribution framework with 6 primary and 15 secondary categories to facilitate in-depth analysis. |
| Outcome: | The proposed framework is based on a dataset specifically designed for error attribution, along with the corresponding scores and feedback. |
Automatic Evaluation of Attribution by Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Generative large language models (LLMs) incorporate external references to generate and support claims. however, evaluating the attribution remains an open problem. |
| Approach: | They investigate automatic evaluation of attribution given by large language models . they define different types of attributed errors and then explore two approaches . |
| Outcome: | The proposed methods highlight promising signals and challenges. |
Talent or Luck? Evaluating Attribution Bias in Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on social biases in large language models focus on surface-level associations or isolated stereotypes. |
| Approach: | They propose a cognitively grounded bias evaluation framework to capture demographic biases across three contexts: single-actor, actor–actor and actor–observer. |
| Outcome: | The proposed framework captures comparative and perspective-driven biases overlooked in previous work. |
A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing evaluations of attribution methods focus on the English language . plausibility and faithfulness are two main criteria for plausible and faithful attributions . |
| Approach: | They propose a cross-lingual strategy to measure faithfulness based on word alignments. |
| Outcome: | The proposed approach eliminates drawbacks of erasure-based evaluations and provides a multilingual dataset with highlights to support future studies. |
On the Interaction of Belief Bias and Explanations (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to evaluate explainability fail to account for belief biases affecting human performance . previous studies have shown that neural models can make confident predictions relying on artifacts . |
| Approach: | They propose to account for belief bias in explainability by using models of varying quality and adversarial examples. |
| Outcome: | The proposed methods show that results change when using models of varying quality and adversarial examples. |
Incorporating Priors with Feature Attribution on Text Classification (P19-1)
Copied to clipboard
| Challenge: | Feature attribution methods are used to help users interpret complex models. |
| Approach: | They propose a feature attribution method that integrates feature attributed features into the objective function to allow machine learning practitioners to incorporate priors in model building. |
| Outcome: | The proposed method reduces undesired model biases without a tradeoff on the original task and improves classifier performance in scarce data setting. |