{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YDFQ24U6244GETZSYZRJ35BE2J","short_pith_number":"pith:YDFQ24U6","canonical_record":{"source":{"id":"2305.12567","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-21T21:06:23Z","cross_cats_sorted":[],"title_canon_sha256":"748f6868493fe0b1f0253a75bc69ac25fc759856c9eaa6fc064cba21d1e860ed","abstract_canon_sha256":"c8eccd48e62d105cf6a8b691e0cf6d37c5fea6010fd9b8a25a4917536caf27b2"},"schema_version":"1.0"},"canonical_sha256":"c0cb0d729ed738624f32c6629df424d25a4daaae6028dd832389a7d37ff0b018","source":{"kind":"arxiv","id":"2305.12567","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.12567","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"arxiv_version","alias_value":"2305.12567v1","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12567","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_12","alias_value":"YDFQ24U6244G","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_16","alias_value":"YDFQ24U6244GETZS","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_8","alias_value":"YDFQ24U6","created_at":"2026-07-05T07:53:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YDFQ24U6244GETZSYZRJ35BE2J","target":"record","payload":{"canonical_record":{"source":{"id":"2305.12567","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-21T21:06:23Z","cross_cats_sorted":[],"title_canon_sha256":"748f6868493fe0b1f0253a75bc69ac25fc759856c9eaa6fc064cba21d1e860ed","abstract_canon_sha256":"c8eccd48e62d105cf6a8b691e0cf6d37c5fea6010fd9b8a25a4917536caf27b2"},"schema_version":"1.0"},"canonical_sha256":"c0cb0d729ed738624f32c6629df424d25a4daaae6028dd832389a7d37ff0b018","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:53:30.183425Z","signature_b64":"Vs/ZgQ4iGITNEboBJBtnH1g0YVcsXzgqmnlHCBl4xlRU2FXh9Qc5OwrC0P+TWUp4UaY+TIi2CDMOyjwh2iR/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0cb0d729ed738624f32c6629df424d25a4daaae6028dd832389a7d37ff0b018","last_reissued_at":"2026-07-05T07:53:30.182944Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:53:30.182944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.12567","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":"2026-07-05T07:53:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hLkRssMkWB37jL0yCo8IjXoev2vu5aaJNaKeaZISHgj2rh/swyNJpGupnU07d9g4G54hxZk9IYlluHGgYvZxAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T19:00:35.309257Z"},"content_sha256":"992c5a44d49b0609e9aa7521908a030d2494a3f1a40da6f93e2fbd5f00ded49b","schema_version":"1.0","event_id":"sha256:992c5a44d49b0609e9aa7521908a030d2494a3f1a40da6f93e2fbd5f00ded49b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YDFQ24U6244GETZSYZRJ35BE2J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Model-Generated Pretraining Signals Improves Zero-Shot Generalization of Text-to-Text Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alvin Cheung, Chenyan Xiong, Jianfeng Gao, Linyuan Gong, Payal Bajaj, XiaoDong Liu, Xia Song, Yiqing Xie","submitted_at":"2023-05-21T21:06:23Z","abstract_excerpt":"This paper explores the effectiveness of model-generated signals in improving zero-shot generalization of text-to-text Transformers such as T5. We study various designs to pretrain T5 using an auxiliary model to construct more challenging token replacements for the main model to denoise. Key aspects under study include the decoding target, the location of the RTD head, and the masking pattern. Based on these studies, we develop a new model, METRO-T0, which is pretrained using the redesigned ELECTRA-Style pretraining strategies and then prompt-finetuned on a mixture of NLP tasks. METRO-T0 outpe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12567","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/2305.12567/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":"2026-07-05T07:53:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UEoJ8IOEc5ekSagZIiQ6FkSwaYy2lI2cIkXi/gQqeP2hZVteO/iLz3bp3WsjDR8K8da6TvXFP+uOPleA80fWDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T19:00:35.309798Z"},"content_sha256":"87c0ae7af34736be65c3a2855d566d11fc0a8bf7565d0ddffbcd8cb4c12d0ffd","schema_version":"1.0","event_id":"sha256:87c0ae7af34736be65c3a2855d566d11fc0a8bf7565d0ddffbcd8cb4c12d0ffd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YDFQ24U6244GETZSYZRJ35BE2J/bundle.json","state_url":"https://pith.science/pith/YDFQ24U6244GETZSYZRJ35BE2J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YDFQ24U6244GETZSYZRJ35BE2J/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":"2026-08-12T19:00:35Z","links":{"resolver":"https://pith.science/pith/YDFQ24U6244GETZSYZRJ35BE2J","bundle":"https://pith.science/pith/YDFQ24U6244GETZSYZRJ35BE2J/bundle.json","state":"https://pith.science/pith/YDFQ24U6244GETZSYZRJ35BE2J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YDFQ24U6244GETZSYZRJ35BE2J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YDFQ24U6244GETZSYZRJ35BE2J","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":"c8eccd48e62d105cf6a8b691e0cf6d37c5fea6010fd9b8a25a4917536caf27b2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-21T21:06:23Z","title_canon_sha256":"748f6868493fe0b1f0253a75bc69ac25fc759856c9eaa6fc064cba21d1e860ed"},"schema_version":"1.0","source":{"id":"2305.12567","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.12567","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"arxiv_version","alias_value":"2305.12567v1","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12567","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_12","alias_value":"YDFQ24U6244G","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_16","alias_value":"YDFQ24U6244GETZS","created_at":"2026-07-05T07:53:30Z"},{"alias_kind":"pith_short_8","alias_value":"YDFQ24U6","created_at":"2026-07-05T07:53:30Z"}],"graph_snapshots":[{"event_id":"sha256:87c0ae7af34736be65c3a2855d566d11fc0a8bf7565d0ddffbcd8cb4c12d0ffd","target":"graph","created_at":"2026-07-05T07:53:30Z","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/2305.12567/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper explores the effectiveness of model-generated signals in improving zero-shot generalization of text-to-text Transformers such as T5. We study various designs to pretrain T5 using an auxiliary model to construct more challenging token replacements for the main model to denoise. Key aspects under study include the decoding target, the location of the RTD head, and the masking pattern. Based on these studies, we develop a new model, METRO-T0, which is pretrained using the redesigned ELECTRA-Style pretraining strategies and then prompt-finetuned on a mixture of NLP tasks. METRO-T0 outpe","authors_text":"Alvin Cheung, Chenyan Xiong, Jianfeng Gao, Linyuan Gong, Payal Bajaj, XiaoDong Liu, Xia Song, Yiqing Xie","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-21T21:06:23Z","title":"Model-Generated Pretraining Signals Improves Zero-Shot Generalization of Text-to-Text Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12567","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":"sha256:992c5a44d49b0609e9aa7521908a030d2494a3f1a40da6f93e2fbd5f00ded49b","target":"record","created_at":"2026-07-05T07:53:30Z","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":"c8eccd48e62d105cf6a8b691e0cf6d37c5fea6010fd9b8a25a4917536caf27b2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-21T21:06:23Z","title_canon_sha256":"748f6868493fe0b1f0253a75bc69ac25fc759856c9eaa6fc064cba21d1e860ed"},"schema_version":"1.0","source":{"id":"2305.12567","kind":"arxiv","version":1}},"canonical_sha256":"c0cb0d729ed738624f32c6629df424d25a4daaae6028dd832389a7d37ff0b018","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c0cb0d729ed738624f32c6629df424d25a4daaae6028dd832389a7d37ff0b018","first_computed_at":"2026-07-05T07:53:30.182944Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:53:30.182944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Vs/ZgQ4iGITNEboBJBtnH1g0YVcsXzgqmnlHCBl4xlRU2FXh9Qc5OwrC0P+TWUp4UaY+TIi2CDMOyjwh2iR/CA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:53:30.183425Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.12567","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:992c5a44d49b0609e9aa7521908a030d2494a3f1a40da6f93e2fbd5f00ded49b","sha256:87c0ae7af34736be65c3a2855d566d11fc0a8bf7565d0ddffbcd8cb4c12d0ffd"],"state_sha256":"e026b4b25c991c5e55bbbc56f1a88461b21b1cd477bf40c1b8d2565af83fdbb0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eRj8NumHgl2Ir0rdV1F+Hm4ne6hxdp7XXr83bGP72r8kitxyZreXnkGfOfj5Q4+2e5zqqEXC0jfKHRyrOyYrDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T19:00:35.334362Z","bundle_sha256":"4116ce302a49119098ee56a03c21749182182ff1f23944fb4ed1d8db158ab95c"}}