{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:6J3FMEQSXDJU5TYJ2V4NOWJJKX","short_pith_number":"pith:6J3FMEQS","canonical_record":{"source":{"id":"2212.00196","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-01T00:53:04Z","cross_cats_sorted":[],"title_canon_sha256":"cb7624d13109dc3edffd4d10a2aead797fe2fa7fba021394da7520a8d24f63d7","abstract_canon_sha256":"4e5f9adb527fa69411a723b0e074f9ac151e2405c87f58c0a013a58ce05c2fbc"},"schema_version":"1.0"},"canonical_sha256":"f276561212b8d34ecf09d578d7592955ee36d56055ba11cd486708cbb19dfffe","source":{"kind":"arxiv","id":"2212.00196","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.00196","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"arxiv_version","alias_value":"2212.00196v2","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.00196","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_12","alias_value":"6J3FMEQSXDJU","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_16","alias_value":"6J3FMEQSXDJU5TYJ","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_8","alias_value":"6J3FMEQS","created_at":"2026-07-05T06:13:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:6J3FMEQSXDJU5TYJ2V4NOWJJKX","target":"record","payload":{"canonical_record":{"source":{"id":"2212.00196","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-01T00:53:04Z","cross_cats_sorted":[],"title_canon_sha256":"cb7624d13109dc3edffd4d10a2aead797fe2fa7fba021394da7520a8d24f63d7","abstract_canon_sha256":"4e5f9adb527fa69411a723b0e074f9ac151e2405c87f58c0a013a58ce05c2fbc"},"schema_version":"1.0"},"canonical_sha256":"f276561212b8d34ecf09d578d7592955ee36d56055ba11cd486708cbb19dfffe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:37.501586Z","signature_b64":"K9D/FrkVQxhi9MtIH/ZWwfZ2QdXLLXbUXpSjhP1nzJ1g9WMvSd0IzuqlPcktqu79cj7xL9y4WGIoJ7KWpYWxDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f276561212b8d34ecf09d578d7592955ee36d56055ba11cd486708cbb19dfffe","last_reissued_at":"2026-07-05T06:13:37.501083Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:37.501083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2212.00196","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":"2026-07-05T06:13:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DCn1+MfRoV1Z4JdDL6jRc0ro8rhum9AtTow8JHnxwDb1UDEZxFiev8lz4fFvIGNnRf1Y87PB45GVeuLqiragAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:03:47.508904Z"},"content_sha256":"c53415b5d1da402e9bacf0de09ac19e113ac1132531f48a0e745a57e4f404fbe","schema_version":"1.0","event_id":"sha256:c53415b5d1da402e9bacf0de09ac19e113ac1132531f48a0e745a57e4f404fbe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:6J3FMEQSXDJU5TYJ2V4NOWJJKX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data-Efficient Finetuning Using Cross-Task Nearest Neighbors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hamish Ivison, Hannaneh Hajishirzi, Noah A. Smith, Pradeep Dasigi","submitted_at":"2022-12-01T00:53:04Z","abstract_excerpt":"Obtaining labeled data to train a model for a task of interest is often expensive. Prior work shows training models on multitask data augmented with task descriptions (prompts) effectively transfers knowledge to new tasks. Towards efficiently building task-specific models, we assume access to a small number (32-1000) of unlabeled target-task examples and use those to retrieve the most similar labeled examples from a large pool of multitask data augmented with prompts. Compared to the current practice of finetuning models on uniformly sampled prompted multitask data (e.g.: FLAN, T0), our approa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.00196","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/2212.00196/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":"2026-07-05T06:13:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H9g/luVuYa1CPW0s9F9LmXW6rbmb9nmxpmA5owm/xWzawM5xkksQ6F66/093EfhhkFshDf6Qtnogrf+u9q3xCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:03:47.509820Z"},"content_sha256":"4a69f9a70b7668f81bc7cf2094ac8fbbdc2daac5c517630013645e61628d41c5","schema_version":"1.0","event_id":"sha256:4a69f9a70b7668f81bc7cf2094ac8fbbdc2daac5c517630013645e61628d41c5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/bundle.json","state_url":"https://pith.science/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/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":"2026-08-08T07:03:47Z","links":{"resolver":"https://pith.science/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX","bundle":"https://pith.science/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/bundle.json","state":"https://pith.science/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6J3FMEQSXDJU5TYJ2V4NOWJJKX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:6J3FMEQSXDJU5TYJ2V4NOWJJKX","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":"4e5f9adb527fa69411a723b0e074f9ac151e2405c87f58c0a013a58ce05c2fbc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-01T00:53:04Z","title_canon_sha256":"cb7624d13109dc3edffd4d10a2aead797fe2fa7fba021394da7520a8d24f63d7"},"schema_version":"1.0","source":{"id":"2212.00196","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.00196","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"arxiv_version","alias_value":"2212.00196v2","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.00196","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_12","alias_value":"6J3FMEQSXDJU","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_16","alias_value":"6J3FMEQSXDJU5TYJ","created_at":"2026-07-05T06:13:37Z"},{"alias_kind":"pith_short_8","alias_value":"6J3FMEQS","created_at":"2026-07-05T06:13:37Z"}],"graph_snapshots":[{"event_id":"sha256:4a69f9a70b7668f81bc7cf2094ac8fbbdc2daac5c517630013645e61628d41c5","target":"graph","created_at":"2026-07-05T06:13:37Z","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/2212.00196/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Obtaining labeled data to train a model for a task of interest is often expensive. Prior work shows training models on multitask data augmented with task descriptions (prompts) effectively transfers knowledge to new tasks. Towards efficiently building task-specific models, we assume access to a small number (32-1000) of unlabeled target-task examples and use those to retrieve the most similar labeled examples from a large pool of multitask data augmented with prompts. Compared to the current practice of finetuning models on uniformly sampled prompted multitask data (e.g.: FLAN, T0), our approa","authors_text":"Hamish Ivison, Hannaneh Hajishirzi, Noah A. Smith, Pradeep Dasigi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-01T00:53:04Z","title":"Data-Efficient Finetuning Using Cross-Task Nearest Neighbors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.00196","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:c53415b5d1da402e9bacf0de09ac19e113ac1132531f48a0e745a57e4f404fbe","target":"record","created_at":"2026-07-05T06:13:37Z","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":"4e5f9adb527fa69411a723b0e074f9ac151e2405c87f58c0a013a58ce05c2fbc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-01T00:53:04Z","title_canon_sha256":"cb7624d13109dc3edffd4d10a2aead797fe2fa7fba021394da7520a8d24f63d7"},"schema_version":"1.0","source":{"id":"2212.00196","kind":"arxiv","version":2}},"canonical_sha256":"f276561212b8d34ecf09d578d7592955ee36d56055ba11cd486708cbb19dfffe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f276561212b8d34ecf09d578d7592955ee36d56055ba11cd486708cbb19dfffe","first_computed_at":"2026-07-05T06:13:37.501083Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:13:37.501083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"K9D/FrkVQxhi9MtIH/ZWwfZ2QdXLLXbUXpSjhP1nzJ1g9WMvSd0IzuqlPcktqu79cj7xL9y4WGIoJ7KWpYWxDw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:13:37.501586Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.00196","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c53415b5d1da402e9bacf0de09ac19e113ac1132531f48a0e745a57e4f404fbe","sha256:4a69f9a70b7668f81bc7cf2094ac8fbbdc2daac5c517630013645e61628d41c5"],"state_sha256":"44b0d0e7f9bda1e1a0f81f819b6ddcd1b96aabecfb1155b893255d6b8b2aa92e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pgt9g740VKQ9cV9oTvWrDVJgVk1eV8iH9kLtazXq+YgjVr/6HyH76kNkNc7+Y46yG8jjEg2evbMk4hJF7UAgBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T07:03:47.516464Z","bundle_sha256":"53c8f2775c7559413363390ff8f6afad461de8761398e8479114c1c988ed0a0d"}}