{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/evaluating-the-faithfulness-of-importance","title":"Evaluating the Faithfulness of Importance Measures in NLP by Recursively Masking Allegedly Important Tokens and Retraining","arxiv_id":"2110.08412","date":"2021-10-15","proceeding":null,"authors":["Andreas Madsen","Nicholas Meade","Vaibhav Adlakha","Siva Reddy"],"abstract":"To explain NLP models a popular approach is to use importance measures, such as attention, which inform input tokens are important for making a prediction. However, an open question is how well these explanations accurately reflect a model's logic, a property called faithfulness. To answer this question, we propose Recursive ROAR, a new faithfulness metric. This works by recursively masking allegedly important tokens and then retraining the model. The principle is that this should result in worse model performance compared to masking random tokens. The result is a performance curve given a masking-ratio. Furthermore, we propose a summarizing metric using relative area-between-curves (RACU), which allows for easy comparison across papers, models, and tasks. We evaluate 4 different importance measures on 8 different datasets, using both LSTM-attention models and RoBERTa models. We find that the faithfulness of importance measures is both model-dependent and task-dependent. This conclusion contradicts previous evaluations in both computer vision and faithfulness of attention literature.","url_abs":"https://arxiv.org/abs/2110.08412v3","url_pdf":"https://arxiv.org/pdf/2110.08412v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"evaluating-the-faithfulness-of-importance","repo_url":"https://github.com/AndreasMadsen/nlp-roar-interpretability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"open-question","task_name":"Open-Ended Question Answering"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roberta","method_name":"RoBERTa"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2110.08412","atlas_url":"https://app.syntology.ai/?focus=2110.08412","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.08412"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/AndreasMadsen/nlp-roar-interpretability","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4},"by_repo_kind":{"official":{"samples":4,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"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"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"711a7575ed8a5d33","entry":"area_between","repo":"AndreasMadsen/nlp-roar-interpretability","repo_kind":"official","path":"export/roar_plot.py","file_url":"https://github.com/AndreasMadsen/nlp-roar-interpretability/blob/HEAD/export/roar_plot.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"711a7575ed8a5d33"}},{"code_sha256_prefix":"30c6d0c1fc44f8df","entry":"format_mean_ci","repo":"AndreasMadsen/nlp-roar-interpretability","repo_kind":"official","path":"export/roar_plot.py","file_url":"https://github.com/AndreasMadsen/nlp-roar-interpretability/blob/HEAD/export/roar_plot.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"30c6d0c1fc44f8df"}},{"code_sha256_prefix":"07f501efc6070e51","entry":"optimal_roar_batch_size","repo":"AndreasMadsen/nlp-roar-interpretability","repo_kind":"official","path":"nlproar/util/optimal_roar_batch_size.py","file_url":"https://github.com/AndreasMadsen/nlp-roar-interpretability/blob/HEAD/nlproar/util/optimal_roar_batch_size.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"07f501efc6070e51"}},{"code_sha256_prefix":"1423b35e902d8a49","entry":"select_test_metric","repo":"AndreasMadsen/nlp-roar-interpretability","repo_kind":"official","path":"export/roar_plot.py","file_url":"https://github.com/AndreasMadsen/nlp-roar-interpretability/blob/HEAD/export/roar_plot.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1423b35e902d8a49"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}