{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:UVL2EWQ6DKLUQ7ZO4B4W2IXVKX","short_pith_number":"pith:UVL2EWQ6","canonical_record":{"source":{"id":"2109.13006","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-09-24T16:19:25Z","cross_cats_sorted":["cs.CL","cs.LG","cs.LO","cs.NE"],"title_canon_sha256":"62de8d8d3073d9c3730c05f319d62ab8539e431b8dc4c5d8bea7237299297c5d","abstract_canon_sha256":"d2e8bab4664ff8dbee7618d7b9713383c361e717cd618e7e2d9f8e8389a43d98"},"schema_version":"1.0"},"canonical_sha256":"a557a25a1e1a97487f2ee0796d22f555f9bcca8b48f3dcf7eb320d4604101a40","source":{"kind":"arxiv","id":"2109.13006","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.13006","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"arxiv_version","alias_value":"2109.13006v1","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.13006","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_12","alias_value":"UVL2EWQ6DKLU","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_16","alias_value":"UVL2EWQ6DKLUQ7ZO","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_8","alias_value":"UVL2EWQ6","created_at":"2026-07-05T03:17:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:UVL2EWQ6DKLUQ7ZO4B4W2IXVKX","target":"record","payload":{"canonical_record":{"source":{"id":"2109.13006","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-09-24T16:19:25Z","cross_cats_sorted":["cs.CL","cs.LG","cs.LO","cs.NE"],"title_canon_sha256":"62de8d8d3073d9c3730c05f319d62ab8539e431b8dc4c5d8bea7237299297c5d","abstract_canon_sha256":"d2e8bab4664ff8dbee7618d7b9713383c361e717cd618e7e2d9f8e8389a43d98"},"schema_version":"1.0"},"canonical_sha256":"a557a25a1e1a97487f2ee0796d22f555f9bcca8b48f3dcf7eb320d4604101a40","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:17:40.137592Z","signature_b64":"IqTIqFcyy5wbYlkGij/i9pwgD6zR4dsnCLsW9cNw70rCSduQInYoPI9cRuCwtWdkOocxYcBFhjp7DPHS7udICw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a557a25a1e1a97487f2ee0796d22f555f9bcca8b48f3dcf7eb320d4604101a40","last_reissued_at":"2026-07-05T03:17:40.136973Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:17:40.136973Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.13006","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:17:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bhdBwHrP3polnlMyc93q9PRP4zuirzrX0RS1vEYCYLtWBHOqS225QdNLbd8n/oWAiCEAs1jhnC/gLLL+ggtBCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:52:28.159109Z"},"content_sha256":"29f2a7c30e9088517c6821544465f49cac6668eb0236e90803fa9634f84dcd8d","schema_version":"1.0","event_id":"sha256:29f2a7c30e9088517c6821544465f49cac6668eb0236e90803fa9634f84dcd8d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:UVL2EWQ6DKLUQ7ZO4B4W2IXVKX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RuleBert: Teaching Soft Rules to Pre-trained Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.LO","cs.NE"],"primary_cat":"cs.AI","authors_text":"Mohammed Saeed, Naser Ahmadi, Paolo Papotti, Preslav Nakov","submitted_at":"2021-09-24T16:19:25Z","abstract_excerpt":"While pre-trained language models (PLMs) are the go-to solution to tackle many natural language processing problems, they are still very limited in their ability to capture and to use common-sense knowledge. In fact, even if information is available in the form of approximate (soft) logical rules, it is not clear how to transfer it to a PLM in order to improve its performance for deductive reasoning tasks. Here, we aim to bridge this gap by teaching PLMs how to reason with soft Horn rules. We introduce a classification task where, given facts and soft rules, the PLM should return a prediction "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.13006","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/2109.13006/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:17:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KSejl9V4muU4EJdupMOTj5Ilkxd/WekI/3PK9YQOVakJufphpjREohNvOrQe9l8LiIF7SaShdWlxfRfi24z4Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:52:28.159631Z"},"content_sha256":"b25f41668164a9d045a233657a4e64766e23d2dfdbc3781011eea743dbf2a85a","schema_version":"1.0","event_id":"sha256:b25f41668164a9d045a233657a4e64766e23d2dfdbc3781011eea743dbf2a85a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/bundle.json","state_url":"https://pith.science/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T15:52:28Z","links":{"resolver":"https://pith.science/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX","bundle":"https://pith.science/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/bundle.json","state":"https://pith.science/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UVL2EWQ6DKLUQ7ZO4B4W2IXVKX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:UVL2EWQ6DKLUQ7ZO4B4W2IXVKX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d2e8bab4664ff8dbee7618d7b9713383c361e717cd618e7e2d9f8e8389a43d98","cross_cats_sorted":["cs.CL","cs.LG","cs.LO","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-09-24T16:19:25Z","title_canon_sha256":"62de8d8d3073d9c3730c05f319d62ab8539e431b8dc4c5d8bea7237299297c5d"},"schema_version":"1.0","source":{"id":"2109.13006","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.13006","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"arxiv_version","alias_value":"2109.13006v1","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.13006","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_12","alias_value":"UVL2EWQ6DKLU","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_16","alias_value":"UVL2EWQ6DKLUQ7ZO","created_at":"2026-07-05T03:17:40Z"},{"alias_kind":"pith_short_8","alias_value":"UVL2EWQ6","created_at":"2026-07-05T03:17:40Z"}],"graph_snapshots":[{"event_id":"sha256:b25f41668164a9d045a233657a4e64766e23d2dfdbc3781011eea743dbf2a85a","target":"graph","created_at":"2026-07-05T03:17:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2109.13006/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While pre-trained language models (PLMs) are the go-to solution to tackle many natural language processing problems, they are still very limited in their ability to capture and to use common-sense knowledge. In fact, even if information is available in the form of approximate (soft) logical rules, it is not clear how to transfer it to a PLM in order to improve its performance for deductive reasoning tasks. Here, we aim to bridge this gap by teaching PLMs how to reason with soft Horn rules. We introduce a classification task where, given facts and soft rules, the PLM should return a prediction ","authors_text":"Mohammed Saeed, Naser Ahmadi, Paolo Papotti, Preslav Nakov","cross_cats":["cs.CL","cs.LG","cs.LO","cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-09-24T16:19:25Z","title":"RuleBert: Teaching Soft Rules to Pre-trained Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.13006","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:29f2a7c30e9088517c6821544465f49cac6668eb0236e90803fa9634f84dcd8d","target":"record","created_at":"2026-07-05T03:17:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d2e8bab4664ff8dbee7618d7b9713383c361e717cd618e7e2d9f8e8389a43d98","cross_cats_sorted":["cs.CL","cs.LG","cs.LO","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-09-24T16:19:25Z","title_canon_sha256":"62de8d8d3073d9c3730c05f319d62ab8539e431b8dc4c5d8bea7237299297c5d"},"schema_version":"1.0","source":{"id":"2109.13006","kind":"arxiv","version":1}},"canonical_sha256":"a557a25a1e1a97487f2ee0796d22f555f9bcca8b48f3dcf7eb320d4604101a40","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a557a25a1e1a97487f2ee0796d22f555f9bcca8b48f3dcf7eb320d4604101a40","first_computed_at":"2026-07-05T03:17:40.136973Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:17:40.136973Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IqTIqFcyy5wbYlkGij/i9pwgD6zR4dsnCLsW9cNw70rCSduQInYoPI9cRuCwtWdkOocxYcBFhjp7DPHS7udICw==","signature_status":"signed_v1","signed_at":"2026-07-05T03:17:40.137592Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.13006","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:29f2a7c30e9088517c6821544465f49cac6668eb0236e90803fa9634f84dcd8d","sha256:b25f41668164a9d045a233657a4e64766e23d2dfdbc3781011eea743dbf2a85a"],"state_sha256":"751d00d299c0ecb09da3dcf241aa2c3e0984006f08e91d9e58fdeeb4b02881d8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aPRPqxRvni5FaA0Vx3QILKVp7nbQXyeQQbGvvZmtuVBd+Rzpv5shoyRNpFUrzL0w1SZrsq4fN8wkZiR/6SQ3BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T15:52:28.163260Z","bundle_sha256":"a9fd4190cca84893f7840fccd9c3bde6d8c001f273ffe0d14bb20c6a6c79f6b1"}}