{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:R356WUSXSC4QNEILV2IUDKTEYS","short_pith_number":"pith:R356WUSX","canonical_record":{"source":{"id":"2412.17032","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-22T14:17:12Z","cross_cats_sorted":[],"title_canon_sha256":"55870cee40caf9bffd87ea84491123ab070b62bf6bea7bfffaf381463645f557","abstract_canon_sha256":"7eb393f02bdda499b49ecfaeb7ab8e3cc2f18e2d7adb1496d35865e3665d2b5e"},"schema_version":"1.0"},"canonical_sha256":"8efbeb525790b906910bae9141aa64c4bd7a563dbcdf1d851bb15f97133cf286","source":{"kind":"arxiv","id":"2412.17032","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.17032","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"arxiv_version","alias_value":"2412.17032v3","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17032","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_12","alias_value":"R356WUSXSC4Q","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_16","alias_value":"R356WUSXSC4QNEIL","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_8","alias_value":"R356WUSX","created_at":"2026-07-05T11:57:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:R356WUSXSC4QNEILV2IUDKTEYS","target":"record","payload":{"canonical_record":{"source":{"id":"2412.17032","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-22T14:17:12Z","cross_cats_sorted":[],"title_canon_sha256":"55870cee40caf9bffd87ea84491123ab070b62bf6bea7bfffaf381463645f557","abstract_canon_sha256":"7eb393f02bdda499b49ecfaeb7ab8e3cc2f18e2d7adb1496d35865e3665d2b5e"},"schema_version":"1.0"},"canonical_sha256":"8efbeb525790b906910bae9141aa64c4bd7a563dbcdf1d851bb15f97133cf286","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:33.656013Z","signature_b64":"uG1dYn+6B3SAOXjzCgi4Sczyinz+SyqQrv8L4WFnXUl81EhXIpIiXHTO2RWsKJiFNau5X+BOfKrisTMKx4rZCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8efbeb525790b906910bae9141aa64c4bd7a563dbcdf1d851bb15f97133cf286","last_reissued_at":"2026-07-05T11:57:33.655481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:33.655481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.17032","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:57:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eO2QmpMl6Azz3fjELJzs5eZfJinT4eqsTO5QZDQ/tGNTU2zK795BI0SRnR6Wh45/npoycZfHBuWmPilvEwOZDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T23:22:07.233210Z"},"content_sha256":"95ccb9c99efc9299ed222e6c73b9f6867141ef4bfbdabfa108ead07c3b049f7d","schema_version":"1.0","event_id":"sha256:95ccb9c99efc9299ed222e6c73b9f6867141ef4bfbdabfa108ead07c3b049f7d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:R356WUSXSC4QNEILV2IUDKTEYS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MINTQA: A Multi-Hop Question Answering Benchmark for Evaluating LLMs on New and Tail Knowledge","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jeff Z. Pan, Jiaoyan Chen, Jie He, Nan Hu, Wanqiu Long","submitted_at":"2024-12-22T14:17:12Z","abstract_excerpt":"Large language models (LLMs) have demonstrated impressive capabilities in various reasoning tasks but face significant challenges with complex, knowledge-intensive multi-hop queries, particularly those involving new or long-tail knowledge. Existing benchmarks often fail to fully address these challenges. To bridge this gap, we introduce MINTQA (Multi-hop Question Answering on New and Tail Knowledge), a comprehensive benchmark to evaluate LLMs' capabilities in multi-hop reasoning across four critical dimensions: question handling strategy, sub-question generation, retrieval-augmented generation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17032","kind":"arxiv","version":3},"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/2412.17032/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":"2026-07-05T11:57:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OTp7NEhTpSQZvTm3/ZZmFs5PMZBt/LPc1JybQcPAlWZpAFzW/s2obkyQ9SjY0BCuuWO2ymxJ++TN9/xi/aARAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T23:22:07.234095Z"},"content_sha256":"cf0ad05f120582bc5a524f5ca6ddb9383df49ebad6db5b4c74570a4965a70688","schema_version":"1.0","event_id":"sha256:cf0ad05f120582bc5a524f5ca6ddb9383df49ebad6db5b4c74570a4965a70688"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R356WUSXSC4QNEILV2IUDKTEYS/bundle.json","state_url":"https://pith.science/pith/R356WUSXSC4QNEILV2IUDKTEYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R356WUSXSC4QNEILV2IUDKTEYS/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":"2026-08-13T23:22:07Z","links":{"resolver":"https://pith.science/pith/R356WUSXSC4QNEILV2IUDKTEYS","bundle":"https://pith.science/pith/R356WUSXSC4QNEILV2IUDKTEYS/bundle.json","state":"https://pith.science/pith/R356WUSXSC4QNEILV2IUDKTEYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R356WUSXSC4QNEILV2IUDKTEYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:R356WUSXSC4QNEILV2IUDKTEYS","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":"7eb393f02bdda499b49ecfaeb7ab8e3cc2f18e2d7adb1496d35865e3665d2b5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-22T14:17:12Z","title_canon_sha256":"55870cee40caf9bffd87ea84491123ab070b62bf6bea7bfffaf381463645f557"},"schema_version":"1.0","source":{"id":"2412.17032","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.17032","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"arxiv_version","alias_value":"2412.17032v3","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17032","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_12","alias_value":"R356WUSXSC4Q","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_16","alias_value":"R356WUSXSC4QNEIL","created_at":"2026-07-05T11:57:33Z"},{"alias_kind":"pith_short_8","alias_value":"R356WUSX","created_at":"2026-07-05T11:57:33Z"}],"graph_snapshots":[{"event_id":"sha256:cf0ad05f120582bc5a524f5ca6ddb9383df49ebad6db5b4c74570a4965a70688","target":"graph","created_at":"2026-07-05T11:57:33Z","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/2412.17032/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated impressive capabilities in various reasoning tasks but face significant challenges with complex, knowledge-intensive multi-hop queries, particularly those involving new or long-tail knowledge. Existing benchmarks often fail to fully address these challenges. To bridge this gap, we introduce MINTQA (Multi-hop Question Answering on New and Tail Knowledge), a comprehensive benchmark to evaluate LLMs' capabilities in multi-hop reasoning across four critical dimensions: question handling strategy, sub-question generation, retrieval-augmented generation","authors_text":"Jeff Z. Pan, Jiaoyan Chen, Jie He, Nan Hu, Wanqiu Long","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-22T14:17:12Z","title":"MINTQA: A Multi-Hop Question Answering Benchmark for Evaluating LLMs on New and Tail Knowledge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17032","kind":"arxiv","version":3},"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":"sha256:95ccb9c99efc9299ed222e6c73b9f6867141ef4bfbdabfa108ead07c3b049f7d","target":"record","created_at":"2026-07-05T11:57:33Z","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":"7eb393f02bdda499b49ecfaeb7ab8e3cc2f18e2d7adb1496d35865e3665d2b5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-22T14:17:12Z","title_canon_sha256":"55870cee40caf9bffd87ea84491123ab070b62bf6bea7bfffaf381463645f557"},"schema_version":"1.0","source":{"id":"2412.17032","kind":"arxiv","version":3}},"canonical_sha256":"8efbeb525790b906910bae9141aa64c4bd7a563dbcdf1d851bb15f97133cf286","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8efbeb525790b906910bae9141aa64c4bd7a563dbcdf1d851bb15f97133cf286","first_computed_at":"2026-07-05T11:57:33.655481Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:57:33.655481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uG1dYn+6B3SAOXjzCgi4Sczyinz+SyqQrv8L4WFnXUl81EhXIpIiXHTO2RWsKJiFNau5X+BOfKrisTMKx4rZCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:57:33.656013Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.17032","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:95ccb9c99efc9299ed222e6c73b9f6867141ef4bfbdabfa108ead07c3b049f7d","sha256:cf0ad05f120582bc5a524f5ca6ddb9383df49ebad6db5b4c74570a4965a70688"],"state_sha256":"aabf37db9b45430b5f3e1baebe3d489873fb7f92ab1fb65ab51df666d6549795"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Edj1tYDVxskUhiRZ4vQTIxePZfbpn2q27KTq3cgKZogeYbwEVV429palcgsqcO7wyR4jub2sPkkYXCtsPQJMBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T23:22:07.294412Z","bundle_sha256":"7e1f13f1c9d592d67fe6be994a8847d944234d2f7aaf905e166dc95410bcded6"}}