{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:KHFDKAIHI32NODLDADLFUJSUCN","short_pith_number":"pith:KHFDKAIH","schema_version":"1.0","canonical_sha256":"51ca35010746f4d70d6300d65a2654135a0913c2e4c4abe322695e141a68f13f","source":{"kind":"arxiv","id":"2205.09067","version":5},"attestation_state":"computed","paper":{"title":"Automatic Rule Induction for Interpretable Semi-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenguang Zhu, Michael Zeng, Reid Pryzant, Yichong Xu, Ziyi Yang","submitted_at":"2022-05-18T16:50:20Z","abstract_excerpt":"Semi-supervised learning has shown promise in allowing NLP models to generalize from small amounts of labeled data. Meanwhile, pretrained transformer models act as black-box correlation engines that are difficult to explain and sometimes behave unreliably. In this paper, we propose tackling both of these challenges via Automatic Rule Induction (ARI), a simple and general-purpose framework for the automatic discovery and integration of symbolic rules into pretrained transformer models. First, we extract weak symbolic rules from low-capacity machine learning models trained on small amounts of la"},"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":"2205.09067","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-18T16:50:20Z","cross_cats_sorted":[],"title_canon_sha256":"ec44d7f7599342e54ed77aeda01bd048eba5b279a6b46199e42562083c73e63a","abstract_canon_sha256":"82fe8a661a83f4adf1a8c795afa549246c6fd0023653a349bc7d4ab254ef570f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:31.372181Z","signature_b64":"7kqb13GzU7+xYIzukA0kc2ZX5aOtma7FIhMik3kfx4V9EIeWOMrcvIwdyYZd0ijhFCJddlWw6h2JkIYuoYhJCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51ca35010746f4d70d6300d65a2654135a0913c2e4c4abe322695e141a68f13f","last_reissued_at":"2026-07-05T05:06:31.371740Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:31.371740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automatic Rule Induction for Interpretable Semi-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenguang Zhu, Michael Zeng, Reid Pryzant, Yichong Xu, Ziyi Yang","submitted_at":"2022-05-18T16:50:20Z","abstract_excerpt":"Semi-supervised learning has shown promise in allowing NLP models to generalize from small amounts of labeled data. Meanwhile, pretrained transformer models act as black-box correlation engines that are difficult to explain and sometimes behave unreliably. In this paper, we propose tackling both of these challenges via Automatic Rule Induction (ARI), a simple and general-purpose framework for the automatic discovery and integration of symbolic rules into pretrained transformer models. First, we extract weak symbolic rules from low-capacity machine learning models trained on small amounts of la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.09067","kind":"arxiv","version":5},"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/2205.09067/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":"2205.09067","created_at":"2026-07-05T05:06:31.371794+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.09067v5","created_at":"2026-07-05T05:06:31.371794+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.09067","created_at":"2026-07-05T05:06:31.371794+00:00"},{"alias_kind":"pith_short_12","alias_value":"KHFDKAIHI32N","created_at":"2026-07-05T05:06:31.371794+00:00"},{"alias_kind":"pith_short_16","alias_value":"KHFDKAIHI32NODLD","created_at":"2026-07-05T05:06:31.371794+00:00"},{"alias_kind":"pith_short_8","alias_value":"KHFDKAIH","created_at":"2026-07-05T05:06:31.371794+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/KHFDKAIHI32NODLDADLFUJSUCN","json":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN.json","graph_json":"https://pith.science/api/pith-number/KHFDKAIHI32NODLDADLFUJSUCN/graph.json","events_json":"https://pith.science/api/pith-number/KHFDKAIHI32NODLDADLFUJSUCN/events.json","paper":"https://pith.science/paper/KHFDKAIH"},"agent_actions":{"view_html":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN","download_json":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN.json","view_paper":"https://pith.science/paper/KHFDKAIH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.09067&json=true","fetch_graph":"https://pith.science/api/pith-number/KHFDKAIHI32NODLDADLFUJSUCN/graph.json","fetch_events":"https://pith.science/api/pith-number/KHFDKAIHI32NODLDADLFUJSUCN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN/action/storage_attestation","attest_author":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN/action/author_attestation","sign_citation":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN/action/citation_signature","submit_replication":"https://pith.science/pith/KHFDKAIHI32NODLDADLFUJSUCN/action/replication_record"}},"created_at":"2026-07-05T05:06:31.371794+00:00","updated_at":"2026-07-05T05:06:31.371794+00:00"}