{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LGNEU5DVCV46SQGRRDIRAQP3GS","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":"3cab90673c2e41d7338497f9265e03c80cf467ad537e407becd9c73ec21b139d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-31T19:07:37Z","title_canon_sha256":"55b4709ca6cf19e7e79cd42ee6a895a2c322aca5f16ec13cfc202e4aaab4a230"},"schema_version":"1.0","source":{"id":"2402.00123","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.00123","created_at":"2026-07-05T09:28:10Z"},{"alias_kind":"arxiv_version","alias_value":"2402.00123v2","created_at":"2026-07-05T09:28:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00123","created_at":"2026-07-05T09:28:10Z"},{"alias_kind":"pith_short_12","alias_value":"LGNEU5DVCV46","created_at":"2026-07-05T09:28:10Z"},{"alias_kind":"pith_short_16","alias_value":"LGNEU5DVCV46SQGR","created_at":"2026-07-05T09:28:10Z"},{"alias_kind":"pith_short_8","alias_value":"LGNEU5DV","created_at":"2026-07-05T09:28:10Z"}],"graph_snapshots":[{"event_id":"sha256:0b19d7c9023cb94168e8077bfabb42ce780d16e2c6efd18a8c58c0f476d4a6c0","target":"graph","created_at":"2026-07-05T09:28:10Z","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/2402.00123/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The differences between cloze-task language model (LM) probing with 1) expert-made templates and 2) naturally-occurring text have often been overlooked. Here, we evaluate 16 different LMs on 10 probing English datasets -- 4 template-based and 6 template-free -- in general and biomedical domains to answer the following research questions: (RQ1) Do model rankings differ between the two approaches? (RQ2) Do models' absolute scores differ between the two approaches? (RQ3) Do the answers to RQ1 and RQ2 differ between general and domain-specific models? Our findings are: 1) Template-free and templat","authors_text":"Katharina von der Wense, Kevin Bennett, Lawrence E Hunter, Sagi Shaier","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-31T19:07:37Z","title":"Comparing Template-based and Template-free Language Model Probing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00123","kind":"arxiv","version":2},"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":"sha256:96b8101a285125af03bbaf5063d36aabebf99328329006c8002fdda1228450e2","target":"record","created_at":"2026-07-05T09:28:10Z","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":"3cab90673c2e41d7338497f9265e03c80cf467ad537e407becd9c73ec21b139d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-31T19:07:37Z","title_canon_sha256":"55b4709ca6cf19e7e79cd42ee6a895a2c322aca5f16ec13cfc202e4aaab4a230"},"schema_version":"1.0","source":{"id":"2402.00123","kind":"arxiv","version":2}},"canonical_sha256":"599a4a74751579e940d188d11041fb3496d5329068740ef98d9f2a56915028d8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"599a4a74751579e940d188d11041fb3496d5329068740ef98d9f2a56915028d8","first_computed_at":"2026-07-05T09:28:10.353605Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:28:10.353605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zmHX5BoanixZSK4hXJP10wEvaxZqyaJ5XaVGaR4ajOe35ljRsaU2M8hG4AQiH8Aect31wgvFowpO+cylJsccDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:28:10.354059Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.00123","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:96b8101a285125af03bbaf5063d36aabebf99328329006c8002fdda1228450e2","sha256:0b19d7c9023cb94168e8077bfabb42ce780d16e2c6efd18a8c58c0f476d4a6c0"],"state_sha256":"a0c39bc1fc985327e99ab005dfded7b55bf5ff33a18faf9c68ec46f89d03b3fb"}