{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:CBQOQGFYS53IEXNPRDRYYUCTAD","short_pith_number":"pith:CBQOQGFY","canonical_record":{"source":{"id":"2004.14287","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-29T16:11:26Z","cross_cats_sorted":[],"title_canon_sha256":"ba868c670ff9cf26db2de97d0bbfb2b97855dec961e86ecc5b76e19813d7fa5a","abstract_canon_sha256":"ff30cf63bbb83f4d31abd7bd96743d8cf3c468dbdd8b0f4ef24e294d326b6549"},"schema_version":"1.0"},"canonical_sha256":"1060e818b89776825daf88e38c505300ebd39b877ba35548ca3b7e9a43318d06","source":{"kind":"arxiv","id":"2004.14287","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.14287","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"arxiv_version","alias_value":"2004.14287v1","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.14287","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_12","alias_value":"CBQOQGFYS53I","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_16","alias_value":"CBQOQGFYS53IEXNP","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_8","alias_value":"CBQOQGFY","created_at":"2026-07-05T00:59:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:CBQOQGFYS53IEXNPRDRYYUCTAD","target":"record","payload":{"canonical_record":{"source":{"id":"2004.14287","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-29T16:11:26Z","cross_cats_sorted":[],"title_canon_sha256":"ba868c670ff9cf26db2de97d0bbfb2b97855dec961e86ecc5b76e19813d7fa5a","abstract_canon_sha256":"ff30cf63bbb83f4d31abd7bd96743d8cf3c468dbdd8b0f4ef24e294d326b6549"},"schema_version":"1.0"},"canonical_sha256":"1060e818b89776825daf88e38c505300ebd39b877ba35548ca3b7e9a43318d06","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:59:19.142367Z","signature_b64":"mJCC+EUs33uCE3wHfTaiPtga6lHMaurMUpildMNJ0i6dYWp1SbbqsZpCiFulMZ1ZTcMR76rK2Ys8bgK52uh9AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1060e818b89776825daf88e38c505300ebd39b877ba35548ca3b7e9a43318d06","last_reissued_at":"2026-07-05T00:59:19.141950Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:59:19.141950Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2004.14287","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":"2026-07-05T00:59:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pvha+alK7lRpKQ9YjVY/CiDvbLo5vD4/S1tEsgrkFYbAxr+8zIvRzVE1VDs/Fq+x76sISHYZ18YiRKpBRhuRBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T20:13:34.811891Z"},"content_sha256":"a4024ada06b4b6aaed404a8d2f88f3f3c53a438b13e4ae4e0885b0fbf9121a03","schema_version":"1.0","event_id":"sha256:a4024ada06b4b6aaed404a8d2f88f3f3c53a438b13e4ae4e0885b0fbf9121a03"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:CBQOQGFYS53IEXNPRDRYYUCTAD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"General Purpose Text Embeddings from Pre-trained Language Models for Scalable Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoran Li, Jingfei Du, Myle Ott, Veselin Stoyanov, Xing Zhou","submitted_at":"2020-04-29T16:11:26Z","abstract_excerpt":"The state of the art on many NLP tasks is currently achieved by large pre-trained language models, which require a considerable amount of computation. We explore a setting where many different predictions are made on a single piece of text. In that case, some of the computational cost during inference can be amortized over the different tasks using a shared text encoder. We compare approaches for training such an encoder and show that encoders pre-trained over multiple tasks generalize well to unseen tasks. We also compare ways of extracting fixed- and limited-size representations from this en"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.14287","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/2004.14287/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":"2026-07-05T00:59:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S/U0nkFoa1Cr0Ta1Nb/B/D97bZZLhCrTmfXLKGJ9T95fkqn8n6CFTcW7DkQMXDjeA96Ne9QuiM3B6iWlZbLODQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T20:13:34.812545Z"},"content_sha256":"98d9f2f2404d1f30825df9baecc674b2f1c16b61ea6efda49e2ff15da0060fe1","schema_version":"1.0","event_id":"sha256:98d9f2f2404d1f30825df9baecc674b2f1c16b61ea6efda49e2ff15da0060fe1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/bundle.json","state_url":"https://pith.science/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/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":"2026-08-05T20:13:34Z","links":{"resolver":"https://pith.science/pith/CBQOQGFYS53IEXNPRDRYYUCTAD","bundle":"https://pith.science/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/bundle.json","state":"https://pith.science/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CBQOQGFYS53IEXNPRDRYYUCTAD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:CBQOQGFYS53IEXNPRDRYYUCTAD","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":"ff30cf63bbb83f4d31abd7bd96743d8cf3c468dbdd8b0f4ef24e294d326b6549","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-29T16:11:26Z","title_canon_sha256":"ba868c670ff9cf26db2de97d0bbfb2b97855dec961e86ecc5b76e19813d7fa5a"},"schema_version":"1.0","source":{"id":"2004.14287","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.14287","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"arxiv_version","alias_value":"2004.14287v1","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.14287","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_12","alias_value":"CBQOQGFYS53I","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_16","alias_value":"CBQOQGFYS53IEXNP","created_at":"2026-07-05T00:59:19Z"},{"alias_kind":"pith_short_8","alias_value":"CBQOQGFY","created_at":"2026-07-05T00:59:19Z"}],"graph_snapshots":[{"event_id":"sha256:98d9f2f2404d1f30825df9baecc674b2f1c16b61ea6efda49e2ff15da0060fe1","target":"graph","created_at":"2026-07-05T00:59:19Z","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/2004.14287/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The state of the art on many NLP tasks is currently achieved by large pre-trained language models, which require a considerable amount of computation. We explore a setting where many different predictions are made on a single piece of text. In that case, some of the computational cost during inference can be amortized over the different tasks using a shared text encoder. We compare approaches for training such an encoder and show that encoders pre-trained over multiple tasks generalize well to unseen tasks. We also compare ways of extracting fixed- and limited-size representations from this en","authors_text":"Haoran Li, Jingfei Du, Myle Ott, Veselin Stoyanov, Xing Zhou","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-29T16:11:26Z","title":"General Purpose Text Embeddings from Pre-trained Language Models for Scalable Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.14287","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:a4024ada06b4b6aaed404a8d2f88f3f3c53a438b13e4ae4e0885b0fbf9121a03","target":"record","created_at":"2026-07-05T00:59:19Z","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":"ff30cf63bbb83f4d31abd7bd96743d8cf3c468dbdd8b0f4ef24e294d326b6549","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-29T16:11:26Z","title_canon_sha256":"ba868c670ff9cf26db2de97d0bbfb2b97855dec961e86ecc5b76e19813d7fa5a"},"schema_version":"1.0","source":{"id":"2004.14287","kind":"arxiv","version":1}},"canonical_sha256":"1060e818b89776825daf88e38c505300ebd39b877ba35548ca3b7e9a43318d06","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1060e818b89776825daf88e38c505300ebd39b877ba35548ca3b7e9a43318d06","first_computed_at":"2026-07-05T00:59:19.141950Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:59:19.141950Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mJCC+EUs33uCE3wHfTaiPtga6lHMaurMUpildMNJ0i6dYWp1SbbqsZpCiFulMZ1ZTcMR76rK2Ys8bgK52uh9AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T00:59:19.142367Z","signed_message":"canonical_sha256_bytes"},"source_id":"2004.14287","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a4024ada06b4b6aaed404a8d2f88f3f3c53a438b13e4ae4e0885b0fbf9121a03","sha256:98d9f2f2404d1f30825df9baecc674b2f1c16b61ea6efda49e2ff15da0060fe1"],"state_sha256":"4ee249dd7d21239c5b9ea354115094af18ec724329c7b86656ea66acc1975af1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J7SJ5PnWKLDQXzCGWz6ovk/duKcXc8+O4GanTgBCVYU9IzzoC4P2cKI2IKofVJIYBIt94NQQbwoPFhUB6Gz8Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T20:13:34.818552Z","bundle_sha256":"2c01763d95523f59338c1531a3e5fdb8430b4bee219f7c02f5b9520d07682e96"}}