{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SEYSV7GLIHUEN43KH4QUXHJI6L","short_pith_number":"pith:SEYSV7GL","canonical_record":{"source":{"id":"2403.09472","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-14T15:12:38Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"a13a1ba7013e6a266252b888bead1dcc8e3e7f47a113c96dee508f11639bc68e","abstract_canon_sha256":"be219a5017f56bbc5396e205f67f58105f16a65b782fe3283d711c28b8b07b41"},"schema_version":"1.0"},"canonical_sha256":"91312afccb41e846f36a3f214b9d28f2f82a209606c7a10e7f752feec556fa56","source":{"kind":"arxiv","id":"2403.09472","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09472","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09472v2","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09472","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_12","alias_value":"SEYSV7GLIHUE","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_16","alias_value":"SEYSV7GLIHUEN43K","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_8","alias_value":"SEYSV7GL","created_at":"2026-07-05T09:46:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SEYSV7GLIHUEN43KH4QUXHJI6L","target":"record","payload":{"canonical_record":{"source":{"id":"2403.09472","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-14T15:12:38Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"a13a1ba7013e6a266252b888bead1dcc8e3e7f47a113c96dee508f11639bc68e","abstract_canon_sha256":"be219a5017f56bbc5396e205f67f58105f16a65b782fe3283d711c28b8b07b41"},"schema_version":"1.0"},"canonical_sha256":"91312afccb41e846f36a3f214b9d28f2f82a209606c7a10e7f752feec556fa56","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:46:45.495365Z","signature_b64":"RgOI2dGvLtHMPPWubg4KL69zie6Zd+XKA+ylB+KxOt0QUn2sMbuizIbwF3wYo6YLxDBB2yLfGI+gKzhrZsMwCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91312afccb41e846f36a3f214b9d28f2f82a209606c7a10e7f752feec556fa56","last_reissued_at":"2026-07-05T09:46:45.494842Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:46:45.494842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.09472","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":"2026-07-05T09:46:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kz2D577svh8AtyEe87JTM0FH3N5MQ/WkbSY9L8GTQOa7yVjN5mOGt77UXaVM1hECMJ6feKLOv1JBWznaOaLsBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T02:04:42.065913Z"},"content_sha256":"4a556a0abaf64fde6928c9a9cf1ee9891d36c84c317c0c7c59559e642de3c596","schema_version":"1.0","event_id":"sha256:4a556a0abaf64fde6928c9a9cf1ee9891d36c84c317c0c7c59559e642de3c596"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SEYSV7GLIHUEN43KH4QUXHJI6L","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Chuang Gan, Longhui Yu, Sean Welleck, Weiyang Liu, Yikang Shen, Yiming Yang, Zhiqing Sun","submitted_at":"2024-03-14T15:12:38Z","abstract_excerpt":"Current AI alignment methodologies rely on human-provided demonstrations or judgments, and the learned capabilities of AI systems would be upper-bounded by human capabilities as a result. This raises a challenging research question: How can we keep improving the systems when their capabilities have surpassed the levels of humans? This paper answers this question in the context of tackling hard reasoning tasks (e.g., level 4-5 MATH problems) via learning from human annotations on easier tasks (e.g., level 1-3 MATH problems), which we term as easy-to-hard generalization. Our key insight is that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09472","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/2403.09472/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":"2026-07-05T09:46:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5ruSzfXhZywaATfLdlQWETIdha819zcxj1ZckuxCBIlhIvAvt0f0LkPd2qWQWucmPEN/uElihjgJiMjZuONBAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T02:04:42.066830Z"},"content_sha256":"961b014c74d92e3e5125a8cc593a2832d6b846c17a63df2100f52588bc439b9c","schema_version":"1.0","event_id":"sha256:961b014c74d92e3e5125a8cc593a2832d6b846c17a63df2100f52588bc439b9c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/bundle.json","state_url":"https://pith.science/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/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":"2026-08-18T02:04:42Z","links":{"resolver":"https://pith.science/pith/SEYSV7GLIHUEN43KH4QUXHJI6L","bundle":"https://pith.science/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/bundle.json","state":"https://pith.science/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SEYSV7GLIHUEN43KH4QUXHJI6L/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SEYSV7GLIHUEN43KH4QUXHJI6L","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":"be219a5017f56bbc5396e205f67f58105f16a65b782fe3283d711c28b8b07b41","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-14T15:12:38Z","title_canon_sha256":"a13a1ba7013e6a266252b888bead1dcc8e3e7f47a113c96dee508f11639bc68e"},"schema_version":"1.0","source":{"id":"2403.09472","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09472","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09472v2","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09472","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_12","alias_value":"SEYSV7GLIHUE","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_16","alias_value":"SEYSV7GLIHUEN43K","created_at":"2026-07-05T09:46:45Z"},{"alias_kind":"pith_short_8","alias_value":"SEYSV7GL","created_at":"2026-07-05T09:46:45Z"}],"graph_snapshots":[{"event_id":"sha256:961b014c74d92e3e5125a8cc593a2832d6b846c17a63df2100f52588bc439b9c","target":"graph","created_at":"2026-07-05T09:46:45Z","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/2403.09472/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current AI alignment methodologies rely on human-provided demonstrations or judgments, and the learned capabilities of AI systems would be upper-bounded by human capabilities as a result. This raises a challenging research question: How can we keep improving the systems when their capabilities have surpassed the levels of humans? This paper answers this question in the context of tackling hard reasoning tasks (e.g., level 4-5 MATH problems) via learning from human annotations on easier tasks (e.g., level 1-3 MATH problems), which we term as easy-to-hard generalization. Our key insight is that ","authors_text":"Chuang Gan, Longhui Yu, Sean Welleck, Weiyang Liu, Yikang Shen, Yiming Yang, Zhiqing Sun","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-14T15:12:38Z","title":"Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09472","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":"sha256:4a556a0abaf64fde6928c9a9cf1ee9891d36c84c317c0c7c59559e642de3c596","target":"record","created_at":"2026-07-05T09:46:45Z","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":"be219a5017f56bbc5396e205f67f58105f16a65b782fe3283d711c28b8b07b41","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-14T15:12:38Z","title_canon_sha256":"a13a1ba7013e6a266252b888bead1dcc8e3e7f47a113c96dee508f11639bc68e"},"schema_version":"1.0","source":{"id":"2403.09472","kind":"arxiv","version":2}},"canonical_sha256":"91312afccb41e846f36a3f214b9d28f2f82a209606c7a10e7f752feec556fa56","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"91312afccb41e846f36a3f214b9d28f2f82a209606c7a10e7f752feec556fa56","first_computed_at":"2026-07-05T09:46:45.494842Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:46:45.494842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RgOI2dGvLtHMPPWubg4KL69zie6Zd+XKA+ylB+KxOt0QUn2sMbuizIbwF3wYo6YLxDBB2yLfGI+gKzhrZsMwCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:46:45.495365Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.09472","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4a556a0abaf64fde6928c9a9cf1ee9891d36c84c317c0c7c59559e642de3c596","sha256:961b014c74d92e3e5125a8cc593a2832d6b846c17a63df2100f52588bc439b9c"],"state_sha256":"52473d45fcba6aecce1138e63122cfb8fb9dc66abcb92dd06481dca4e12545d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Sl3RWq6qYrsEJ/2Eooba8fUYyKPgmJugB5CnfSINL2G8cpxqicAqMtvT805QUrEwptIWpb1ogr3aPNwWnGOjBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T02:04:42.075151Z","bundle_sha256":"eac45c1ed655249fd8edc0ad2caef9873618e17cc539cd6bac84dccdf5f74168"}}