{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LYZJZUUHOEYNPHPBQT4F6A2QQF","short_pith_number":"pith:LYZJZUUH","canonical_record":{"source":{"id":"2501.02683","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T23:19:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"21bb68984ef1c5150c278ce4c49dd2c461e6edafb6410068da3aa14e311a931d","abstract_canon_sha256":"f4d3bfd9374aff21ab4109b3219b74456b35c75aaad9f16dae2e2413d296e860"},"schema_version":"1.0"},"canonical_sha256":"5e329cd2877130d79de184f85f03508144689190b6ae446788017965164773b7","source":{"kind":"arxiv","id":"2501.02683","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.02683","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"arxiv_version","alias_value":"2501.02683v2","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02683","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_12","alias_value":"LYZJZUUHOEYN","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_16","alias_value":"LYZJZUUHOEYNPHPB","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_8","alias_value":"LYZJZUUH","created_at":"2026-07-05T09:58:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LYZJZUUHOEYNPHPBQT4F6A2QQF","target":"record","payload":{"canonical_record":{"source":{"id":"2501.02683","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T23:19:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"21bb68984ef1c5150c278ce4c49dd2c461e6edafb6410068da3aa14e311a931d","abstract_canon_sha256":"f4d3bfd9374aff21ab4109b3219b74456b35c75aaad9f16dae2e2413d296e860"},"schema_version":"1.0"},"canonical_sha256":"5e329cd2877130d79de184f85f03508144689190b6ae446788017965164773b7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:58:27.733992Z","signature_b64":"Lakjc9PLhr89rJKEaRsjMdpQNTWbSdLXQKz9LkZcV+GPuivg7e0die8oCuSbWbssDowUYwgT06QtAg7ulPV6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e329cd2877130d79de184f85f03508144689190b6ae446788017965164773b7","last_reissued_at":"2026-07-05T09:58:27.733591Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:58:27.733591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.02683","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:58:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5R4aq0H0ogvOlRrrO+nSgpFsYdFOcCzmjoJb6fWMcM9NCF3iaCpkXjLL2LJPJb4+aKGz9z7lwmwyIhL6OKpiAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T12:22:10.257711Z"},"content_sha256":"db6d3fb2978096da16be9840e4e9e844b17293a1697c440afa8b741bbfdbd487","schema_version":"1.0","event_id":"sha256:db6d3fb2978096da16be9840e4e9e844b17293a1697c440afa8b741bbfdbd487"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LYZJZUUHOEYNPHPBQT4F6A2QQF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Superficial Patterns to Semantic Understanding: Fine-Tuning Language Models on Contrast Sets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Daniel Petrov","submitted_at":"2025-01-05T23:19:55Z","abstract_excerpt":"Large-scale pre-trained language models have demonstrated high performance on standard datasets for natural language inference (NLI) tasks. Unfortunately, these evaluations can be misleading, as although the models can perform well on in-distribution data, they perform poorly on out-of-distribution test sets, such as contrast sets. Contrast sets consist of perturbed instances of data that have very minor, but meaningful, changes to the input that alter the gold label, revealing how models can learn superficial patterns in the training data rather than learning more sophisticated language nuanc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02683","kind":"arxiv","version":2},"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/2501.02683/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":"2026-07-05T09:58:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GvZT6BQ64KI6k8E7p3tNZQFXLsrartV+pVyzLv9fs5II9goAnnnJbhovqS9d+o8imxtJ5NJXiVEbSwQ9HBf9DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T12:22:10.258388Z"},"content_sha256":"73366474cd8ada30fa66504aba31eda8d3485e7a87a9f260ea7a65fd7a845e96","schema_version":"1.0","event_id":"sha256:73366474cd8ada30fa66504aba31eda8d3485e7a87a9f260ea7a65fd7a845e96"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/bundle.json","state_url":"https://pith.science/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/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":"2026-08-13T12:22:10Z","links":{"resolver":"https://pith.science/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF","bundle":"https://pith.science/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/bundle.json","state":"https://pith.science/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LYZJZUUHOEYNPHPBQT4F6A2QQF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LYZJZUUHOEYNPHPBQT4F6A2QQF","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":"f4d3bfd9374aff21ab4109b3219b74456b35c75aaad9f16dae2e2413d296e860","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T23:19:55Z","title_canon_sha256":"21bb68984ef1c5150c278ce4c49dd2c461e6edafb6410068da3aa14e311a931d"},"schema_version":"1.0","source":{"id":"2501.02683","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.02683","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"arxiv_version","alias_value":"2501.02683v2","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02683","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_12","alias_value":"LYZJZUUHOEYN","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_16","alias_value":"LYZJZUUHOEYNPHPB","created_at":"2026-07-05T09:58:27Z"},{"alias_kind":"pith_short_8","alias_value":"LYZJZUUH","created_at":"2026-07-05T09:58:27Z"}],"graph_snapshots":[{"event_id":"sha256:73366474cd8ada30fa66504aba31eda8d3485e7a87a9f260ea7a65fd7a845e96","target":"graph","created_at":"2026-07-05T09:58:27Z","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/2501.02683/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-scale pre-trained language models have demonstrated high performance on standard datasets for natural language inference (NLI) tasks. Unfortunately, these evaluations can be misleading, as although the models can perform well on in-distribution data, they perform poorly on out-of-distribution test sets, such as contrast sets. Contrast sets consist of perturbed instances of data that have very minor, but meaningful, changes to the input that alter the gold label, revealing how models can learn superficial patterns in the training data rather than learning more sophisticated language nuanc","authors_text":"Daniel Petrov","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T23:19:55Z","title":"From Superficial Patterns to Semantic Understanding: Fine-Tuning Language Models on Contrast Sets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02683","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:db6d3fb2978096da16be9840e4e9e844b17293a1697c440afa8b741bbfdbd487","target":"record","created_at":"2026-07-05T09:58:27Z","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":"f4d3bfd9374aff21ab4109b3219b74456b35c75aaad9f16dae2e2413d296e860","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T23:19:55Z","title_canon_sha256":"21bb68984ef1c5150c278ce4c49dd2c461e6edafb6410068da3aa14e311a931d"},"schema_version":"1.0","source":{"id":"2501.02683","kind":"arxiv","version":2}},"canonical_sha256":"5e329cd2877130d79de184f85f03508144689190b6ae446788017965164773b7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5e329cd2877130d79de184f85f03508144689190b6ae446788017965164773b7","first_computed_at":"2026-07-05T09:58:27.733591Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:58:27.733591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Lakjc9PLhr89rJKEaRsjMdpQNTWbSdLXQKz9LkZcV+GPuivg7e0die8oCuSbWbssDowUYwgT06QtAg7ulPV6CA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:58:27.733992Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.02683","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:db6d3fb2978096da16be9840e4e9e844b17293a1697c440afa8b741bbfdbd487","sha256:73366474cd8ada30fa66504aba31eda8d3485e7a87a9f260ea7a65fd7a845e96"],"state_sha256":"abed86c6fb93f1261e1fe911b1c3aee7018885fa250e1a0ef05cdba1dfe79b5c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/gEg8vJr9S/N3ytQnU9mG18v5GPyI6roXzvcCrZiv876dRkiMs+ykNYmbecxfsJcVOwRvu3THP6nMMRhtKDcBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T12:22:10.265562Z","bundle_sha256":"bf6a99ff4ea3d78fe02ba8d85d07cc45b30f78de39487d0f50b253364a58da61"}}