{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:ZNIW5RJHLEWFQSUNE7IFRJL7VA","short_pith_number":"pith:ZNIW5RJH","canonical_record":{"source":{"id":"2010.07835","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-15T15:55:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"588bf2cb2c51325b4722a8d697f7904dc3d86e6f266cfbcb4207a2f494f9e77d","abstract_canon_sha256":"d0237959644b0fbaf830e8198bbcf12c9c8e366e9e034b8000f01a82ee411361"},"schema_version":"1.0"},"canonical_sha256":"cb516ec527592c584a8d27d058a57fa813ee38b267db159d05ad71e78f5e869f","source":{"kind":"arxiv","id":"2010.07835","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.07835","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"arxiv_version","alias_value":"2010.07835v3","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.07835","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_12","alias_value":"ZNIW5RJHLEWF","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_16","alias_value":"ZNIW5RJHLEWFQSUN","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_8","alias_value":"ZNIW5RJH","created_at":"2026-07-05T02:27:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:ZNIW5RJHLEWFQSUNE7IFRJL7VA","target":"record","payload":{"canonical_record":{"source":{"id":"2010.07835","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-15T15:55:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"588bf2cb2c51325b4722a8d697f7904dc3d86e6f266cfbcb4207a2f494f9e77d","abstract_canon_sha256":"d0237959644b0fbaf830e8198bbcf12c9c8e366e9e034b8000f01a82ee411361"},"schema_version":"1.0"},"canonical_sha256":"cb516ec527592c584a8d27d058a57fa813ee38b267db159d05ad71e78f5e869f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:27:54.852220Z","signature_b64":"fSHROVu2htIZqTEXsorNqhI0vO8QBciqyHgLMtKkNn8/fnmZAaw1jnZGFl7fqFh74PM8sw+8oGh+u19QACw7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb516ec527592c584a8d27d058a57fa813ee38b267db159d05ad71e78f5e869f","last_reissued_at":"2026-07-05T02:27:54.851790Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:27:54.851790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.07835","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:27:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7zLy7kaKenr0Q9Ff31crh6iMT+MPyrtIjKfLkZKWtmNHW+xdESCXlYQhePjC02WAuX2eKTFK6AJ/G2HTL5xrDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T13:42:06.165899Z"},"content_sha256":"a0a797c9506b66d53dc51d392d854d85b007535e8a54d40d6ee630c95ff4603b","schema_version":"1.0","event_id":"sha256:a0a797c9506b66d53dc51d392d854d85b007535e8a54d40d6ee630c95ff4603b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:ZNIW5RJHLEWFQSUNE7IFRJL7VA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Zhang, Haoming Jiang, Simiao Zuo, Tuo Zhao, Wendi Ren, Yue Yu","submitted_at":"2020-10-15T15:55:08Z","abstract_excerpt":"Fine-tuned pre-trained language models (LMs) have achieved enormous success in many natural language processing (NLP) tasks, but they still require excessive labeled data in the fine-tuning stage. We study the problem of fine-tuning pre-trained LMs using only weak supervision, without any labeled data. This problem is challenging because the high capacity of LMs makes them prone to overfitting the noisy labels generated by weak supervision. To address this problem, we develop a contrastive self-training framework, COSINE, to enable fine-tuning LMs with weak supervision. Underpinned by contrast"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.07835","kind":"arxiv","version":3},"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/2010.07835/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":"2026-07-05T02:27:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mRV5myTFr/OizcWHZVe4lSudtTPPSLEOcEY1eoukVLlbx4c/2UAhdCr95a9uL7DVDkHkbVdKCaqoElTimjNLDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T13:42:06.166271Z"},"content_sha256":"50e29b4a9d15bd30580d1d262ba2904ebcdad4eeeac866686101c19b1c0a956d","schema_version":"1.0","event_id":"sha256:50e29b4a9d15bd30580d1d262ba2904ebcdad4eeeac866686101c19b1c0a956d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/bundle.json","state_url":"https://pith.science/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/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":"2026-07-07T13:42:06Z","links":{"resolver":"https://pith.science/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA","bundle":"https://pith.science/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/bundle.json","state":"https://pith.science/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZNIW5RJHLEWFQSUNE7IFRJL7VA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:ZNIW5RJHLEWFQSUNE7IFRJL7VA","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":"d0237959644b0fbaf830e8198bbcf12c9c8e366e9e034b8000f01a82ee411361","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-15T15:55:08Z","title_canon_sha256":"588bf2cb2c51325b4722a8d697f7904dc3d86e6f266cfbcb4207a2f494f9e77d"},"schema_version":"1.0","source":{"id":"2010.07835","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.07835","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"arxiv_version","alias_value":"2010.07835v3","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.07835","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_12","alias_value":"ZNIW5RJHLEWF","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_16","alias_value":"ZNIW5RJHLEWFQSUN","created_at":"2026-07-05T02:27:54Z"},{"alias_kind":"pith_short_8","alias_value":"ZNIW5RJH","created_at":"2026-07-05T02:27:54Z"}],"graph_snapshots":[{"event_id":"sha256:50e29b4a9d15bd30580d1d262ba2904ebcdad4eeeac866686101c19b1c0a956d","target":"graph","created_at":"2026-07-05T02:27:54Z","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/2010.07835/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuned pre-trained language models (LMs) have achieved enormous success in many natural language processing (NLP) tasks, but they still require excessive labeled data in the fine-tuning stage. We study the problem of fine-tuning pre-trained LMs using only weak supervision, without any labeled data. This problem is challenging because the high capacity of LMs makes them prone to overfitting the noisy labels generated by weak supervision. To address this problem, we develop a contrastive self-training framework, COSINE, to enable fine-tuning LMs with weak supervision. Underpinned by contrast","authors_text":"Chao Zhang, Haoming Jiang, Simiao Zuo, Tuo Zhao, Wendi Ren, Yue Yu","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-15T15:55:08Z","title":"Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.07835","kind":"arxiv","version":3},"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:a0a797c9506b66d53dc51d392d854d85b007535e8a54d40d6ee630c95ff4603b","target":"record","created_at":"2026-07-05T02:27:54Z","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":"d0237959644b0fbaf830e8198bbcf12c9c8e366e9e034b8000f01a82ee411361","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-15T15:55:08Z","title_canon_sha256":"588bf2cb2c51325b4722a8d697f7904dc3d86e6f266cfbcb4207a2f494f9e77d"},"schema_version":"1.0","source":{"id":"2010.07835","kind":"arxiv","version":3}},"canonical_sha256":"cb516ec527592c584a8d27d058a57fa813ee38b267db159d05ad71e78f5e869f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb516ec527592c584a8d27d058a57fa813ee38b267db159d05ad71e78f5e869f","first_computed_at":"2026-07-05T02:27:54.851790Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:27:54.851790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fSHROVu2htIZqTEXsorNqhI0vO8QBciqyHgLMtKkNn8/fnmZAaw1jnZGFl7fqFh74PM8sw+8oGh+u19QACw7Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:27:54.852220Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.07835","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a0a797c9506b66d53dc51d392d854d85b007535e8a54d40d6ee630c95ff4603b","sha256:50e29b4a9d15bd30580d1d262ba2904ebcdad4eeeac866686101c19b1c0a956d"],"state_sha256":"08571ea7ab1d80962025d6ca206cf608f241d082964d68a5eb6ed9fe70451f8d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Jv+Nj+PpIHQeslHBf10bAkJIru47S5R5DMYMFKoWKJ6gCtO5+bKouSOZNuJaQp9k0ZfHm2WQF2eZbLEzkS53BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T13:42:06.168236Z","bundle_sha256":"e3be5c38db2a258f31730a261837f2d38381f2451ec1177b3620438a910b6b67"}}