{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5C4BVOFWLJXLTY3HVOT5VGFG7F","short_pith_number":"pith:5C4BVOFW","schema_version":"1.0","canonical_sha256":"e8b81ab8b65a6eb9e367aba7da98a6f950e3b915360a90a9f1f995d8131cdbb1","source":{"kind":"arxiv","id":"2305.19689","version":1},"attestation_state":"computed","paper":{"title":"Assessing Word Importance Using Models Trained for Semantic Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"D\\'avid Javorsk\\'y, Fran\\c{c}ois Yvon, Ond\\v{r}ej Bojar","submitted_at":"2023-05-31T09:34:26Z","abstract_excerpt":"Many NLP tasks require to automatically identify the most significant words in a text. In this work, we derive word significance from models trained to solve semantic task: Natural Language Inference and Paraphrase Identification. Using an attribution method aimed to explain the predictions of these models, we derive importance scores for each input token. We evaluate their relevance using a so-called cross-task evaluation: Analyzing the performance of one model on an input masked according to the other model's weight, we show that our method is robust with respect to the choice of the initial"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2305.19689","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-31T09:34:26Z","cross_cats_sorted":[],"title_canon_sha256":"6f5b373ad1d0cb8904b7eca6967d367fe3899959f080f753f61eec8cb38af309","abstract_canon_sha256":"45a153cd5e5b5f457a8762b97e6675613e5b8ebd68a26d987430a05e821b6418"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:15:53.814006Z","signature_b64":"EMzzYisJ/Ve0xW2s0gTSTWYsG+4iYYflZ1ovVC+1/0ivo+O/58A3T9BZQMqLyRynoM4O163xtZHhl0uJm6/3Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8b81ab8b65a6eb9e367aba7da98a6f950e3b915360a90a9f1f995d8131cdbb1","last_reissued_at":"2026-07-05T06:15:53.813632Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:15:53.813632Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Assessing Word Importance Using Models Trained for Semantic Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"D\\'avid Javorsk\\'y, Fran\\c{c}ois Yvon, Ond\\v{r}ej Bojar","submitted_at":"2023-05-31T09:34:26Z","abstract_excerpt":"Many NLP tasks require to automatically identify the most significant words in a text. In this work, we derive word significance from models trained to solve semantic task: Natural Language Inference and Paraphrase Identification. Using an attribution method aimed to explain the predictions of these models, we derive importance scores for each input token. We evaluate their relevance using a so-called cross-task evaluation: Analyzing the performance of one model on an input masked according to the other model's weight, we show that our method is robust with respect to the choice of the initial"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.19689","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2305.19689/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2305.19689","created_at":"2026-07-05T06:15:53.813690+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.19689v1","created_at":"2026-07-05T06:15:53.813690+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.19689","created_at":"2026-07-05T06:15:53.813690+00:00"},{"alias_kind":"pith_short_12","alias_value":"5C4BVOFWLJXL","created_at":"2026-07-05T06:15:53.813690+00:00"},{"alias_kind":"pith_short_16","alias_value":"5C4BVOFWLJXLTY3H","created_at":"2026-07-05T06:15:53.813690+00:00"},{"alias_kind":"pith_short_8","alias_value":"5C4BVOFW","created_at":"2026-07-05T06:15:53.813690+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F","json":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F.json","graph_json":"https://pith.science/api/pith-number/5C4BVOFWLJXLTY3HVOT5VGFG7F/graph.json","events_json":"https://pith.science/api/pith-number/5C4BVOFWLJXLTY3HVOT5VGFG7F/events.json","paper":"https://pith.science/paper/5C4BVOFW"},"agent_actions":{"view_html":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F","download_json":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F.json","view_paper":"https://pith.science/paper/5C4BVOFW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.19689&json=true","fetch_graph":"https://pith.science/api/pith-number/5C4BVOFWLJXLTY3HVOT5VGFG7F/graph.json","fetch_events":"https://pith.science/api/pith-number/5C4BVOFWLJXLTY3HVOT5VGFG7F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F/action/storage_attestation","attest_author":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F/action/author_attestation","sign_citation":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F/action/citation_signature","submit_replication":"https://pith.science/pith/5C4BVOFWLJXLTY3HVOT5VGFG7F/action/replication_record"}},"created_at":"2026-07-05T06:15:53.813690+00:00","updated_at":"2026-07-05T06:15:53.813690+00:00"}