{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:7HHJWTOFA4ACACLD73VT4PCGLY","short_pith_number":"pith:7HHJWTOF","canonical_record":{"source":{"id":"2501.11877","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T04:11:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"50bff10f8abf0d90225190f57494340b87619863be6b8052a092b6253dcfb442","abstract_canon_sha256":"1cff7b676726edd079021eec5c189e189e8142144745da865bcd573e2ff4009a"},"schema_version":"1.0"},"canonical_sha256":"f9ce9b4dc50700200963feeb3e3c465e36e963c4d849dfa43cd8df11c2fafd5c","source":{"kind":"arxiv","id":"2501.11877","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.11877","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"arxiv_version","alias_value":"2501.11877v1","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11877","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_12","alias_value":"7HHJWTOFA4AC","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_16","alias_value":"7HHJWTOFA4ACACLD","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_8","alias_value":"7HHJWTOF","created_at":"2026-07-05T10:03:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:7HHJWTOFA4ACACLD73VT4PCGLY","target":"record","payload":{"canonical_record":{"source":{"id":"2501.11877","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T04:11:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"50bff10f8abf0d90225190f57494340b87619863be6b8052a092b6253dcfb442","abstract_canon_sha256":"1cff7b676726edd079021eec5c189e189e8142144745da865bcd573e2ff4009a"},"schema_version":"1.0"},"canonical_sha256":"f9ce9b4dc50700200963feeb3e3c465e36e963c4d849dfa43cd8df11c2fafd5c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:15.272536Z","signature_b64":"rdwof0ZvW8gKRqI6Q43+jszO1vDHCuZPWNGFzGWDRB1zS0E4+4waneT52i19BnrlVbZKiE/s9Ty2WeqZksvGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9ce9b4dc50700200963feeb3e3c465e36e963c4d849dfa43cd8df11c2fafd5c","last_reissued_at":"2026-07-05T10:03:15.272047Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:15.272047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.11877","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":"2026-07-05T10:03:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pHg8xl+I8fJM1Xeb9tluv+ugXOCwH2ajNMkEaWDxu+qTQQJpjKFxt0tYkFGvSRfJSxtq7kMhVU/vjQ+KilDEAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:24:43.050867Z"},"content_sha256":"4718752d57f7b7550872ba804e3f65e948a46f21b32a569847ed7b63b8b05e26","schema_version":"1.0","event_id":"sha256:4718752d57f7b7550872ba804e3f65e948a46f21b32a569847ed7b63b8b05e26"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:7HHJWTOFA4ACACLD73VT4PCGLY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Drafts to Answers: Unlocking LLM Potential via Aggregation Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ganqu Cui, Sen Yang, Tingchen Fu, Yafu Li, Yu Cheng, Zhilin Wang","submitted_at":"2025-01-21T04:11:59Z","abstract_excerpt":"Scaling data and model size has been proven effective for boosting the performance of large language models. In addition to training-time scaling, recent studies have revealed that increasing test-time computational resources can further improve performance. In this work, we introduce Aggregation Fine-Tuning (AFT), a supervised finetuning paradigm where the model learns to synthesize multiple draft responses, referred to as proposals, into a single, refined answer, termed aggregation. At inference time, a propose-and-aggregate strategy further boosts performance by iteratively generating propo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11877","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/2501.11877/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":"2026-07-05T10:03:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Br3wuKYNMDNIDWur8e/aSsa8ECmfKckBXnwBXxuX5X6isaDnuYxfUVSDg47bSei8r3czWv4UF+tc4bDuH0OdAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:24:43.051388Z"},"content_sha256":"e3d8f869168afc58985bdc5551416a46e28417c87c3418777b85ca2d562cb680","schema_version":"1.0","event_id":"sha256:e3d8f869168afc58985bdc5551416a46e28417c87c3418777b85ca2d562cb680"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7HHJWTOFA4ACACLD73VT4PCGLY/bundle.json","state_url":"https://pith.science/pith/7HHJWTOFA4ACACLD73VT4PCGLY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7HHJWTOFA4ACACLD73VT4PCGLY/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":"2026-08-06T05:24:43Z","links":{"resolver":"https://pith.science/pith/7HHJWTOFA4ACACLD73VT4PCGLY","bundle":"https://pith.science/pith/7HHJWTOFA4ACACLD73VT4PCGLY/bundle.json","state":"https://pith.science/pith/7HHJWTOFA4ACACLD73VT4PCGLY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7HHJWTOFA4ACACLD73VT4PCGLY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7HHJWTOFA4ACACLD73VT4PCGLY","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":"1cff7b676726edd079021eec5c189e189e8142144745da865bcd573e2ff4009a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T04:11:59Z","title_canon_sha256":"50bff10f8abf0d90225190f57494340b87619863be6b8052a092b6253dcfb442"},"schema_version":"1.0","source":{"id":"2501.11877","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.11877","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"arxiv_version","alias_value":"2501.11877v1","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11877","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_12","alias_value":"7HHJWTOFA4AC","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_16","alias_value":"7HHJWTOFA4ACACLD","created_at":"2026-07-05T10:03:15Z"},{"alias_kind":"pith_short_8","alias_value":"7HHJWTOF","created_at":"2026-07-05T10:03:15Z"}],"graph_snapshots":[{"event_id":"sha256:e3d8f869168afc58985bdc5551416a46e28417c87c3418777b85ca2d562cb680","target":"graph","created_at":"2026-07-05T10:03:15Z","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/2501.11877/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Scaling data and model size has been proven effective for boosting the performance of large language models. In addition to training-time scaling, recent studies have revealed that increasing test-time computational resources can further improve performance. In this work, we introduce Aggregation Fine-Tuning (AFT), a supervised finetuning paradigm where the model learns to synthesize multiple draft responses, referred to as proposals, into a single, refined answer, termed aggregation. At inference time, a propose-and-aggregate strategy further boosts performance by iteratively generating propo","authors_text":"Ganqu Cui, Sen Yang, Tingchen Fu, Yafu Li, Yu Cheng, Zhilin Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T04:11:59Z","title":"From Drafts to Answers: Unlocking LLM Potential via Aggregation Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11877","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":"sha256:4718752d57f7b7550872ba804e3f65e948a46f21b32a569847ed7b63b8b05e26","target":"record","created_at":"2026-07-05T10:03:15Z","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":"1cff7b676726edd079021eec5c189e189e8142144745da865bcd573e2ff4009a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T04:11:59Z","title_canon_sha256":"50bff10f8abf0d90225190f57494340b87619863be6b8052a092b6253dcfb442"},"schema_version":"1.0","source":{"id":"2501.11877","kind":"arxiv","version":1}},"canonical_sha256":"f9ce9b4dc50700200963feeb3e3c465e36e963c4d849dfa43cd8df11c2fafd5c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f9ce9b4dc50700200963feeb3e3c465e36e963c4d849dfa43cd8df11c2fafd5c","first_computed_at":"2026-07-05T10:03:15.272047Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:03:15.272047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rdwof0ZvW8gKRqI6Q43+jszO1vDHCuZPWNGFzGWDRB1zS0E4+4waneT52i19BnrlVbZKiE/s9Ty2WeqZksvGCg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:03:15.272536Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.11877","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4718752d57f7b7550872ba804e3f65e948a46f21b32a569847ed7b63b8b05e26","sha256:e3d8f869168afc58985bdc5551416a46e28417c87c3418777b85ca2d562cb680"],"state_sha256":"b416673cf2a56c180d2bbc9b09e3ebaa0dbd9bcab3fc3b2e0984777f82b866a1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DQTtylCsutZ0SkWhB39/A7mT2EPgVVk0zABaJgjlufU9TjOXkIuvVehB2/3Sjw/O+3djzZaLgjuXTBgcLblwCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T05:24:43.057287Z","bundle_sha256":"d55b4db8c7ddda2f8329f01af12d923b81784110ff2fd69212528b030e9e517d"}}