{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:5MKRY4XG34FZKXXVFXAEJHJ2VA","short_pith_number":"pith:5MKRY4XG","canonical_record":{"source":{"id":"2408.16737","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-29T17:32:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fbbc81d0390a198333eb52194f609649e9c62a7a4476265e21f6d153aebf565a","abstract_canon_sha256":"43c939145b47a50f58b83d98b2b39265df5594e85ac3f8751f981605a23b4c36"},"schema_version":"1.0"},"canonical_sha256":"eb151c72e6df0b955ef52dc0449d3aa838991f6c82452f71350dd14256a0dddf","source":{"kind":"arxiv","id":"2408.16737","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.16737","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"arxiv_version","alias_value":"2408.16737v2","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.16737","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_12","alias_value":"5MKRY4XG34FZ","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_16","alias_value":"5MKRY4XG34FZKXXV","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_8","alias_value":"5MKRY4XG","created_at":"2026-07-05T09:17:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:5MKRY4XG34FZKXXVFXAEJHJ2VA","target":"record","payload":{"canonical_record":{"source":{"id":"2408.16737","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-29T17:32:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fbbc81d0390a198333eb52194f609649e9c62a7a4476265e21f6d153aebf565a","abstract_canon_sha256":"43c939145b47a50f58b83d98b2b39265df5594e85ac3f8751f981605a23b4c36"},"schema_version":"1.0"},"canonical_sha256":"eb151c72e6df0b955ef52dc0449d3aa838991f6c82452f71350dd14256a0dddf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:15.079202Z","signature_b64":"NHZvdSIwpfqNq2AIehTO1LZsVYtdYBmQbafLUT+zLBe1QiUOxxveJ40zLFLaK2/HsQ1TVPXRmcD4FSnU8b7VDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eb151c72e6df0b955ef52dc0449d3aa838991f6c82452f71350dd14256a0dddf","last_reissued_at":"2026-07-05T09:17:15.078769Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:15.078769Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.16737","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":"2026-07-05T09:17:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"naSh5Rd3vIrkTJ7oX6OTZTMgdeIJ81Fw6eOkvcu/P6Wmt2epsOVyhin31/XzH2688cOYkt0rpXc6YcoO2ydfBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:26:13.214248Z"},"content_sha256":"f2f2e9fdbb6f2d469bdd94120cde7aa82ab95f0aeb379eff49589c26c40d723a","schema_version":"1.0","event_id":"sha256:f2f2e9fdbb6f2d469bdd94120cde7aa82ab95f0aeb379eff49589c26c40d723a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:5MKRY4XG34FZKXXVFXAEJHJ2VA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Smaller, Weaker, Yet Better: Training LLM Reasoners via Compute-Optimal Sampling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arian Hosseini, Hritik Bansal, Mehran Kazemi, Rishabh Agarwal, Vinh Q. Tran","submitted_at":"2024-08-29T17:32:35Z","abstract_excerpt":"Training on high-quality synthetic data from strong language models (LMs) is a common strategy to improve the reasoning performance of LMs. In this work, we revisit whether this strategy is compute-optimal under a fixed inference budget (e.g., FLOPs). To do so, we investigate the trade-offs between generating synthetic data using a stronger but more expensive (SE) model versus a weaker but cheaper (WC) model. We evaluate the generated data across three key metrics: coverage, diversity, and false positive rate, and show that the data from WC models may have higher coverage and diversity, but al"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.16737","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/2408.16737/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":"2026-07-05T09:17:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/M+/z5fkopm0rBxpfrIpMqs15064HGek/rLh3ehY9HjlOFKQ9fjkfNjLPSmtkqDDzbX5Wx0BC4DzSwzRUatcCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:26:13.214636Z"},"content_sha256":"9671c0d0b08e8753d44e583d9f9451ce653c7e465557f76cdd49e859310ab818","schema_version":"1.0","event_id":"sha256:9671c0d0b08e8753d44e583d9f9451ce653c7e465557f76cdd49e859310ab818"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/bundle.json","state_url":"https://pith.science/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/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":"2026-07-07T09:26:13Z","links":{"resolver":"https://pith.science/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA","bundle":"https://pith.science/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/bundle.json","state":"https://pith.science/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5MKRY4XG34FZKXXVFXAEJHJ2VA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5MKRY4XG34FZKXXVFXAEJHJ2VA","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":"43c939145b47a50f58b83d98b2b39265df5594e85ac3f8751f981605a23b4c36","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-29T17:32:35Z","title_canon_sha256":"fbbc81d0390a198333eb52194f609649e9c62a7a4476265e21f6d153aebf565a"},"schema_version":"1.0","source":{"id":"2408.16737","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.16737","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"arxiv_version","alias_value":"2408.16737v2","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.16737","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_12","alias_value":"5MKRY4XG34FZ","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_16","alias_value":"5MKRY4XG34FZKXXV","created_at":"2026-07-05T09:17:15Z"},{"alias_kind":"pith_short_8","alias_value":"5MKRY4XG","created_at":"2026-07-05T09:17:15Z"}],"graph_snapshots":[{"event_id":"sha256:9671c0d0b08e8753d44e583d9f9451ce653c7e465557f76cdd49e859310ab818","target":"graph","created_at":"2026-07-05T09:17:15Z","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/2408.16737/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training on high-quality synthetic data from strong language models (LMs) is a common strategy to improve the reasoning performance of LMs. In this work, we revisit whether this strategy is compute-optimal under a fixed inference budget (e.g., FLOPs). To do so, we investigate the trade-offs between generating synthetic data using a stronger but more expensive (SE) model versus a weaker but cheaper (WC) model. We evaluate the generated data across three key metrics: coverage, diversity, and false positive rate, and show that the data from WC models may have higher coverage and diversity, but al","authors_text":"Arian Hosseini, Hritik Bansal, Mehran Kazemi, Rishabh Agarwal, Vinh Q. Tran","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-29T17:32:35Z","title":"Smaller, Weaker, Yet Better: Training LLM Reasoners via Compute-Optimal Sampling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.16737","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:f2f2e9fdbb6f2d469bdd94120cde7aa82ab95f0aeb379eff49589c26c40d723a","target":"record","created_at":"2026-07-05T09:17:15Z","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":"43c939145b47a50f58b83d98b2b39265df5594e85ac3f8751f981605a23b4c36","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-29T17:32:35Z","title_canon_sha256":"fbbc81d0390a198333eb52194f609649e9c62a7a4476265e21f6d153aebf565a"},"schema_version":"1.0","source":{"id":"2408.16737","kind":"arxiv","version":2}},"canonical_sha256":"eb151c72e6df0b955ef52dc0449d3aa838991f6c82452f71350dd14256a0dddf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eb151c72e6df0b955ef52dc0449d3aa838991f6c82452f71350dd14256a0dddf","first_computed_at":"2026-07-05T09:17:15.078769Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:17:15.078769Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NHZvdSIwpfqNq2AIehTO1LZsVYtdYBmQbafLUT+zLBe1QiUOxxveJ40zLFLaK2/HsQ1TVPXRmcD4FSnU8b7VDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:17:15.079202Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.16737","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2f2e9fdbb6f2d469bdd94120cde7aa82ab95f0aeb379eff49589c26c40d723a","sha256:9671c0d0b08e8753d44e583d9f9451ce653c7e465557f76cdd49e859310ab818"],"state_sha256":"9e64912d24e5f54ae46543f9ae7f5186ccc11c8b1dd011df9425193c6a6a196b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RUWSVb0EzpgPJK0hJHD/3SJoxaikAdaUH7pBm9iJDflnKtKO6Uc3DtGOLO0yWw0VzCDIBmIvkG32V2vAbl5eAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T09:26:13.216589Z","bundle_sha256":"74f387b15021700a0b0fde3676334a50cbcecd84f528e0505805201cef40fdb1"}}