{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:ECBOWV2LLNCRTHOQU7LIUR2VNU","short_pith_number":"pith:ECBOWV2L","canonical_record":{"source":{"id":"2303.01580","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-02T21:13:56Z","cross_cats_sorted":[],"title_canon_sha256":"516b49ba210236a76d4d4fb2d26298c2022f6e2e697f64a21f7d56a0e35eb70d","abstract_canon_sha256":"99d7e7e330d242a1475154d3dee201c3178c992861ed0bdb3a28d1ecf21a55a9"},"schema_version":"1.0"},"canonical_sha256":"2082eb574b5b45199dd0a7d68a47556d3e4dfac8faffdd6b7b52dd46b5e91e68","source":{"kind":"arxiv","id":"2303.01580","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.01580","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"arxiv_version","alias_value":"2303.01580v2","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.01580","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_12","alias_value":"ECBOWV2LLNCR","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_16","alias_value":"ECBOWV2LLNCRTHOQ","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_8","alias_value":"ECBOWV2L","created_at":"2026-07-05T07:01:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:ECBOWV2LLNCRTHOQU7LIUR2VNU","target":"record","payload":{"canonical_record":{"source":{"id":"2303.01580","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-02T21:13:56Z","cross_cats_sorted":[],"title_canon_sha256":"516b49ba210236a76d4d4fb2d26298c2022f6e2e697f64a21f7d56a0e35eb70d","abstract_canon_sha256":"99d7e7e330d242a1475154d3dee201c3178c992861ed0bdb3a28d1ecf21a55a9"},"schema_version":"1.0"},"canonical_sha256":"2082eb574b5b45199dd0a7d68a47556d3e4dfac8faffdd6b7b52dd46b5e91e68","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:01:59.306752Z","signature_b64":"fo94XVqATFa30T5wI00xChDyesVFHpmpJ1tLzf7fJ5gk3VNJmbtE33Gjra9ZTWPTuo1hLsFPCwhrxQue15kOAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2082eb574b5b45199dd0a7d68a47556d3e4dfac8faffdd6b7b52dd46b5e91e68","last_reissued_at":"2026-07-05T07:01:59.306134Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:01:59.306134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.01580","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":"2026-07-05T07:01:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"We4Vf4DIk+WCKmpK38YmHRpCMfgxaCjHQJ9qKMm+l4Ws12m9y1kfiM93qyP3KxnjqJdr2ByoQBWh7NfalOQ7CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-21T10:31:11.565077Z"},"content_sha256":"3a2ef56bba9b433f683791faac2ce7e2eab28a60b32fa40532c37ce42c6f22ac","schema_version":"1.0","event_id":"sha256:3a2ef56bba9b433f683791faac2ce7e2eab28a60b32fa40532c37ce42c6f22ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:ECBOWV2LLNCRTHOQU7LIUR2VNU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mixture of Soft Prompts for Controllable Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Celine Lee, Derek Chen, Domenic Rosati, Yunan Lu, Zhou Yu","submitted_at":"2023-03-02T21:13:56Z","abstract_excerpt":"Large language models (LLMs) effectively generate fluent text when the target output follows natural language patterns. However, structured prediction tasks confine the output format to a limited ontology, causing even very large models to struggle since they were never trained with such restrictions in mind. The difficulty of using LLMs for direct prediction is exacerbated in few-shot learning scenarios, which commonly arise due to domain shift and resource limitations. We flip the problem on its head by leveraging the LLM as a tool for data augmentation rather than direct prediction. Our pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.01580","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/2303.01580/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":"2026-07-05T07:01:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0DNkBWgkD8bNOBNgY6b9vQ2uBiZZ7I0L2mEIUH63NFPHZiLk20JMVHDqJJQ12130AAYKQlJtpwX3EAIGNE10Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-21T10:31:11.565469Z"},"content_sha256":"e511564883bc7f5f6ba7e0389fc2fe13d7e751f8d7f50dc760f0a8f52505b551","schema_version":"1.0","event_id":"sha256:e511564883bc7f5f6ba7e0389fc2fe13d7e751f8d7f50dc760f0a8f52505b551"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/bundle.json","state_url":"https://pith.science/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/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":"2026-07-21T10:31:11Z","links":{"resolver":"https://pith.science/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU","bundle":"https://pith.science/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/bundle.json","state":"https://pith.science/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ECBOWV2LLNCRTHOQU7LIUR2VNU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ECBOWV2LLNCRTHOQU7LIUR2VNU","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":"99d7e7e330d242a1475154d3dee201c3178c992861ed0bdb3a28d1ecf21a55a9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-02T21:13:56Z","title_canon_sha256":"516b49ba210236a76d4d4fb2d26298c2022f6e2e697f64a21f7d56a0e35eb70d"},"schema_version":"1.0","source":{"id":"2303.01580","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.01580","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"arxiv_version","alias_value":"2303.01580v2","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.01580","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_12","alias_value":"ECBOWV2LLNCR","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_16","alias_value":"ECBOWV2LLNCRTHOQ","created_at":"2026-07-05T07:01:59Z"},{"alias_kind":"pith_short_8","alias_value":"ECBOWV2L","created_at":"2026-07-05T07:01:59Z"}],"graph_snapshots":[{"event_id":"sha256:e511564883bc7f5f6ba7e0389fc2fe13d7e751f8d7f50dc760f0a8f52505b551","target":"graph","created_at":"2026-07-05T07:01:59Z","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/2303.01580/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) effectively generate fluent text when the target output follows natural language patterns. However, structured prediction tasks confine the output format to a limited ontology, causing even very large models to struggle since they were never trained with such restrictions in mind. The difficulty of using LLMs for direct prediction is exacerbated in few-shot learning scenarios, which commonly arise due to domain shift and resource limitations. We flip the problem on its head by leveraging the LLM as a tool for data augmentation rather than direct prediction. Our pro","authors_text":"Celine Lee, Derek Chen, Domenic Rosati, Yunan Lu, Zhou Yu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-02T21:13:56Z","title":"Mixture of Soft Prompts for Controllable Data Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.01580","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":"sha256:3a2ef56bba9b433f683791faac2ce7e2eab28a60b32fa40532c37ce42c6f22ac","target":"record","created_at":"2026-07-05T07:01:59Z","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":"99d7e7e330d242a1475154d3dee201c3178c992861ed0bdb3a28d1ecf21a55a9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-02T21:13:56Z","title_canon_sha256":"516b49ba210236a76d4d4fb2d26298c2022f6e2e697f64a21f7d56a0e35eb70d"},"schema_version":"1.0","source":{"id":"2303.01580","kind":"arxiv","version":2}},"canonical_sha256":"2082eb574b5b45199dd0a7d68a47556d3e4dfac8faffdd6b7b52dd46b5e91e68","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2082eb574b5b45199dd0a7d68a47556d3e4dfac8faffdd6b7b52dd46b5e91e68","first_computed_at":"2026-07-05T07:01:59.306134Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:01:59.306134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fo94XVqATFa30T5wI00xChDyesVFHpmpJ1tLzf7fJ5gk3VNJmbtE33Gjra9ZTWPTuo1hLsFPCwhrxQue15kOAA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:01:59.306752Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.01580","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a2ef56bba9b433f683791faac2ce7e2eab28a60b32fa40532c37ce42c6f22ac","sha256:e511564883bc7f5f6ba7e0389fc2fe13d7e751f8d7f50dc760f0a8f52505b551"],"state_sha256":"4d6ba8792fa0d2dd035a5c4bbfdda27bae015e864d4c8a8688b055f760ae804b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cbmGDbT9JnKMwYX5Sfr60M4/2yl+MbZvPjXYYQuhRef0dbVCMrs9xJhM7MPV+8XVE+T24MPz/YSGO5MtC7GZBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-21T10:31:11.567608Z","bundle_sha256":"29bb073260f2c0e6a72647985057f392bb4e22ab4ad5e1f20d5ad69e719a0ba1"}}