{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:I4DGHULBUTOM3UHCW6HW2J7LII","short_pith_number":"pith:I4DGHULB","canonical_record":{"source":{"id":"2103.10385","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-18T17:13:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fc6b781da8302fbde12766804a6d0e849efd13b802713fdf5c7d46089e73c5f8","abstract_canon_sha256":"18324fdae5d67f1a648e80265b47619493f176ca35731fb66d5533928ac91eb4"},"schema_version":"1.0"},"canonical_sha256":"470663d161a4dccdd0e2b78f6d27eb42303a775c9d9a9424ba44203ec470e162","source":{"kind":"arxiv","id":"2103.10385","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.10385","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2103.10385v2","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.10385","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"I4DGHULBUTOM","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"I4DGHULBUTOM3UHC","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"I4DGHULB","created_at":"2026-07-05T07:04:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:I4DGHULBUTOM3UHCW6HW2J7LII","target":"record","payload":{"canonical_record":{"source":{"id":"2103.10385","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-18T17:13:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fc6b781da8302fbde12766804a6d0e849efd13b802713fdf5c7d46089e73c5f8","abstract_canon_sha256":"18324fdae5d67f1a648e80265b47619493f176ca35731fb66d5533928ac91eb4"},"schema_version":"1.0"},"canonical_sha256":"470663d161a4dccdd0e2b78f6d27eb42303a775c9d9a9424ba44203ec470e162","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:04:40.505972Z","signature_b64":"ROSY44l2emV2KIfh8dWm5SGOJlSGFmbVhkZbgLrFCBhlMtadKDuYItqT0s9p/Pj/sM0eXtrr4bGgpS0Mlk04Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"470663d161a4dccdd0e2b78f6d27eb42303a775c9d9a9424ba44203ec470e162","last_reissued_at":"2026-07-05T07:04:40.505441Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:04:40.505441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2103.10385","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":"2026-07-05T07:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HMihTwWRhDjdV0T7RWmL58rF8MzHYuB0GT7e7cTAVQI2Q09zlABncIJlc82alGUDKrZB/YesQV1DhSEMuFOIDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:18:00.886690Z"},"content_sha256":"25a34c12bf09869fb01f58edf1d665484c4914b2f28da7743ba7d792e1adf37d","schema_version":"1.0","event_id":"sha256:25a34c12bf09869fb01f58edf1d665484c4914b2f28da7743ba7d792e1adf37d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:I4DGHULBUTOM3UHCW6HW2J7LII","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GPT Understands, Too","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jie Tang, Ming Ding, Xiao Liu, Yanan Zheng, Yujie Qian, Zhengxiao Du, Zhilin Yang","submitted_at":"2021-03-18T17:13:50Z","abstract_excerpt":"Prompting a pretrained language model with natural language patterns has been proved effective for natural language understanding (NLU). However, our preliminary study reveals that manual discrete prompts often lead to unstable performance -- e.g., changing a single word in the prompt might result in substantial performance drop. We propose a novel method P-Tuning that employs trainable continuous prompt embeddings in concatenation with discrete prompts. Empirically, P-Tuning not only stabilizes training by minimizing the gap between various discrete prompts, but also improves performance by a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.10385","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/2103.10385/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":"2026-07-05T07:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TXUxSC3hvPFX2lOZV2e0GwzvQ/zd4SBg7Y5NuTqU2aByK09QGo7TV/ikpk6JcvH7EgW5TR+RSeMwPzkEREWXAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:18:00.887470Z"},"content_sha256":"ef55b3f1e3547c01f06bb8af19eeac4fd475dced10447a09cfba7f27e5de80f9","schema_version":"1.0","event_id":"sha256:ef55b3f1e3547c01f06bb8af19eeac4fd475dced10447a09cfba7f27e5de80f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/I4DGHULBUTOM3UHCW6HW2J7LII/bundle.json","state_url":"https://pith.science/pith/I4DGHULBUTOM3UHCW6HW2J7LII/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/I4DGHULBUTOM3UHCW6HW2J7LII/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":"2026-08-06T06:18:00Z","links":{"resolver":"https://pith.science/pith/I4DGHULBUTOM3UHCW6HW2J7LII","bundle":"https://pith.science/pith/I4DGHULBUTOM3UHCW6HW2J7LII/bundle.json","state":"https://pith.science/pith/I4DGHULBUTOM3UHCW6HW2J7LII/state.json","well_known_bundle":"https://pith.science/.well-known/pith/I4DGHULBUTOM3UHCW6HW2J7LII/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:I4DGHULBUTOM3UHCW6HW2J7LII","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":"18324fdae5d67f1a648e80265b47619493f176ca35731fb66d5533928ac91eb4","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-18T17:13:50Z","title_canon_sha256":"fc6b781da8302fbde12766804a6d0e849efd13b802713fdf5c7d46089e73c5f8"},"schema_version":"1.0","source":{"id":"2103.10385","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.10385","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2103.10385v2","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.10385","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"I4DGHULBUTOM","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"I4DGHULBUTOM3UHC","created_at":"2026-07-05T07:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"I4DGHULB","created_at":"2026-07-05T07:04:40Z"}],"graph_snapshots":[{"event_id":"sha256:ef55b3f1e3547c01f06bb8af19eeac4fd475dced10447a09cfba7f27e5de80f9","target":"graph","created_at":"2026-07-05T07:04:40Z","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/2103.10385/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Prompting a pretrained language model with natural language patterns has been proved effective for natural language understanding (NLU). However, our preliminary study reveals that manual discrete prompts often lead to unstable performance -- e.g., changing a single word in the prompt might result in substantial performance drop. We propose a novel method P-Tuning that employs trainable continuous prompt embeddings in concatenation with discrete prompts. Empirically, P-Tuning not only stabilizes training by minimizing the gap between various discrete prompts, but also improves performance by a","authors_text":"Jie Tang, Ming Ding, Xiao Liu, Yanan Zheng, Yujie Qian, Zhengxiao Du, Zhilin Yang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-18T17:13:50Z","title":"GPT Understands, Too"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.10385","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":"sha256:25a34c12bf09869fb01f58edf1d665484c4914b2f28da7743ba7d792e1adf37d","target":"record","created_at":"2026-07-05T07:04:40Z","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":"18324fdae5d67f1a648e80265b47619493f176ca35731fb66d5533928ac91eb4","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-18T17:13:50Z","title_canon_sha256":"fc6b781da8302fbde12766804a6d0e849efd13b802713fdf5c7d46089e73c5f8"},"schema_version":"1.0","source":{"id":"2103.10385","kind":"arxiv","version":2}},"canonical_sha256":"470663d161a4dccdd0e2b78f6d27eb42303a775c9d9a9424ba44203ec470e162","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"470663d161a4dccdd0e2b78f6d27eb42303a775c9d9a9424ba44203ec470e162","first_computed_at":"2026-07-05T07:04:40.505441Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:04:40.505441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ROSY44l2emV2KIfh8dWm5SGOJlSGFmbVhkZbgLrFCBhlMtadKDuYItqT0s9p/Pj/sM0eXtrr4bGgpS0Mlk04Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:04:40.505972Z","signed_message":"canonical_sha256_bytes"},"source_id":"2103.10385","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:25a34c12bf09869fb01f58edf1d665484c4914b2f28da7743ba7d792e1adf37d","sha256:ef55b3f1e3547c01f06bb8af19eeac4fd475dced10447a09cfba7f27e5de80f9"],"state_sha256":"ab51352ed3e1f4f1eca6dc143a602c3c8d19884aee1d2c33fce3c114cab3d8ef"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DonqPfa47MdfNyApEWwOtjUPINxBXReELZOn08HJcMbIJEz9LMmsIsufac+zd1WhhTnVbNkwl1mXTgkblLgRAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T06:18:00.890784Z","bundle_sha256":"9c7559748c62afd9d333de9a7e918ee7a9339ad2a6a36eb732fe4df793400f88"}}