{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:NMNRHNFRF5J3BSAHVQTPDEIEG2","short_pith_number":"pith:NMNRHNFR","canonical_record":{"source":{"id":"2507.01297","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T02:35:47Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"658fa7752fd3945baa8bd0a2a6c945a1f8b60ad3e7a40b5efb1eac8f59c49b4d","abstract_canon_sha256":"b1a2d8685417495dc9910a0982c8cfb65acee2827a159a43aae4a29e682c9d71"},"schema_version":"1.0"},"canonical_sha256":"6b1b13b4b12f53b0c807ac26f1910436ac2d82a9be5ad9f0dc09463df60cccda","source":{"kind":"arxiv","id":"2507.01297","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.01297","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"arxiv_version","alias_value":"2507.01297v2","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01297","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_12","alias_value":"NMNRHNFRF5J3","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"NMNRHNFRF5J3BSAH","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"NMNRHNFR","created_at":"2026-07-05T11:32:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:NMNRHNFRF5J3BSAHVQTPDEIEG2","target":"record","payload":{"canonical_record":{"source":{"id":"2507.01297","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T02:35:47Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"658fa7752fd3945baa8bd0a2a6c945a1f8b60ad3e7a40b5efb1eac8f59c49b4d","abstract_canon_sha256":"b1a2d8685417495dc9910a0982c8cfb65acee2827a159a43aae4a29e682c9d71"},"schema_version":"1.0"},"canonical_sha256":"6b1b13b4b12f53b0c807ac26f1910436ac2d82a9be5ad9f0dc09463df60cccda","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:33.234965Z","signature_b64":"Tb307YnuaZm6f++PVdcsTg6H6GulZMYEJ52lEc5RlkAGlFIxLzv+pffrW0XdXLxxlzgDFErYeWWMVKGswBczCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b1b13b4b12f53b0c807ac26f1910436ac2d82a9be5ad9f0dc09463df60cccda","last_reissued_at":"2026-07-05T11:32:33.234504Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:33.234504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.01297","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":"2026-07-05T11:32:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JGlhdkNKTvAAz6gy7ZgvHz3CCoWp79qSPgkA6+pHbp4T6Iapq7gg+PEU2cUXZBsGOz0dOcmnN1tQlevCdS+RDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:54:25.861945Z"},"content_sha256":"902350a52bde76293da3277808a8a4bc44c27da70952b3b88e808884a8c1c44d","schema_version":"1.0","event_id":"sha256:902350a52bde76293da3277808a8a4bc44c27da70952b3b88e808884a8c1c44d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:NMNRHNFRF5J3BSAHVQTPDEIEG2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Frustratingly Simple Retrieval Improves Challenging, Reasoning-Intensive Benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Michael Duan, Pang Wei Koh, Rulin Shao, Sewon Min, Xinxi Lyu","submitted_at":"2025-07-02T02:35:47Z","abstract_excerpt":"Retrieval-augmented Generation (RAG) has primarily been studied in limited settings, such as factoid question answering; more challenging, reasoning-intensive benchmarks have seen limited success from minimal RAG. In this work, we challenge this prevailing view on established, reasoning-intensive benchmarks: MMLU, MMLU Pro, AGI Eval, GPQA, and MATH. We identify a key missing component in prior work: a usable, web-scale datastore aligned with the breadth of pretraining data. To this end, we introduce CompactDS: a diverse, high-quality, web-scale datastore that achieves high retrieval accuracy a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01297","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/2507.01297/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":"2026-07-05T11:32:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w2niO1La5SOiF08KpzDeMSd4OzyUirDu5tmv535FGZqnwfI/I5DjDBXtY/MPOMTpgGkOWzcidbQp3udSwEVmAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:54:25.863063Z"},"content_sha256":"7dd3734bd4f09c8b3030e0e812f565f91d3a96564d091cbf0d06bc31dabda01a","schema_version":"1.0","event_id":"sha256:7dd3734bd4f09c8b3030e0e812f565f91d3a96564d091cbf0d06bc31dabda01a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/bundle.json","state_url":"https://pith.science/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/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":"2026-08-11T01:54:25Z","links":{"resolver":"https://pith.science/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2","bundle":"https://pith.science/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/bundle.json","state":"https://pith.science/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NMNRHNFRF5J3BSAHVQTPDEIEG2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NMNRHNFRF5J3BSAHVQTPDEIEG2","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":"b1a2d8685417495dc9910a0982c8cfb65acee2827a159a43aae4a29e682c9d71","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T02:35:47Z","title_canon_sha256":"658fa7752fd3945baa8bd0a2a6c945a1f8b60ad3e7a40b5efb1eac8f59c49b4d"},"schema_version":"1.0","source":{"id":"2507.01297","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.01297","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"arxiv_version","alias_value":"2507.01297v2","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01297","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_12","alias_value":"NMNRHNFRF5J3","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"NMNRHNFRF5J3BSAH","created_at":"2026-07-05T11:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"NMNRHNFR","created_at":"2026-07-05T11:32:33Z"}],"graph_snapshots":[{"event_id":"sha256:7dd3734bd4f09c8b3030e0e812f565f91d3a96564d091cbf0d06bc31dabda01a","target":"graph","created_at":"2026-07-05T11:32:33Z","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/2507.01297/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Retrieval-augmented Generation (RAG) has primarily been studied in limited settings, such as factoid question answering; more challenging, reasoning-intensive benchmarks have seen limited success from minimal RAG. In this work, we challenge this prevailing view on established, reasoning-intensive benchmarks: MMLU, MMLU Pro, AGI Eval, GPQA, and MATH. We identify a key missing component in prior work: a usable, web-scale datastore aligned with the breadth of pretraining data. To this end, we introduce CompactDS: a diverse, high-quality, web-scale datastore that achieves high retrieval accuracy a","authors_text":"Michael Duan, Pang Wei Koh, Rulin Shao, Sewon Min, Xinxi Lyu","cross_cats":["cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T02:35:47Z","title":"Frustratingly Simple Retrieval Improves Challenging, Reasoning-Intensive Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01297","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":"sha256:902350a52bde76293da3277808a8a4bc44c27da70952b3b88e808884a8c1c44d","target":"record","created_at":"2026-07-05T11:32:33Z","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":"b1a2d8685417495dc9910a0982c8cfb65acee2827a159a43aae4a29e682c9d71","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-02T02:35:47Z","title_canon_sha256":"658fa7752fd3945baa8bd0a2a6c945a1f8b60ad3e7a40b5efb1eac8f59c49b4d"},"schema_version":"1.0","source":{"id":"2507.01297","kind":"arxiv","version":2}},"canonical_sha256":"6b1b13b4b12f53b0c807ac26f1910436ac2d82a9be5ad9f0dc09463df60cccda","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6b1b13b4b12f53b0c807ac26f1910436ac2d82a9be5ad9f0dc09463df60cccda","first_computed_at":"2026-07-05T11:32:33.234504Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:32:33.234504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Tb307YnuaZm6f++PVdcsTg6H6GulZMYEJ52lEc5RlkAGlFIxLzv+pffrW0XdXLxxlzgDFErYeWWMVKGswBczCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:32:33.234965Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.01297","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:902350a52bde76293da3277808a8a4bc44c27da70952b3b88e808884a8c1c44d","sha256:7dd3734bd4f09c8b3030e0e812f565f91d3a96564d091cbf0d06bc31dabda01a"],"state_sha256":"8c42b1426ad9236d7aea473063996f03b868fb106a6c054c0dc82cbdd187a69b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wlS95dv75Dxzf+WWFkvaEi5lecKqMHZjYoa1BPxJeTRdFLbob0xWAl1h3FxpEgdyepX86nqIEIbQEcz+vWLGDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T01:54:25.868727Z","bundle_sha256":"1e8166ab97cba134270622cf2dc53c6655cbdd265874c3d724d76dd6a6237305"}}