{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FTPH3EWGCCYVJA4B7WPRL7CLIP","short_pith_number":"pith:FTPH3EWG","schema_version":"1.0","canonical_sha256":"2cde7d92c610b1548381fd9f15fc4b43da164e8fa20b1ae17a2f0ee85e52294d","source":{"kind":"arxiv","id":"2409.02465","version":2},"attestation_state":"computed","paper":{"title":"DetectiveQA: Evaluating Long-Context Reasoning on Detective Novels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiasheng Ye, Linlin Li, Qipeng Guo, Qun Liu, Tianxiang Sun, Xiangyang Liu, Xiaoran Liu, Xipeng Qiu, Xuanjing Huang, Zhe Xu, Zhigeng Liu","submitted_at":"2024-09-04T06:28:22Z","abstract_excerpt":"Recently, significant efforts have been devoted to enhancing the long-context capabilities of Large Language Models (LLMs), particularly in long-context reasoning. To facilitate this research, we propose \\textbf{DetectiveQA}, a dataset specifically designed for narrative reasoning within long contexts. We leverage detective novels, averaging over 100k tokens, to create a dataset containing 1200 human-annotated questions in both Chinese and English, each paired with corresponding reference reasoning steps. Furthermore, we introduce a step-wise reasoning metric, which enhances the evaluation of "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2409.02465","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-04T06:28:22Z","cross_cats_sorted":[],"title_canon_sha256":"91c0f9e1e3d44c3ffb09cc21cec001e401ca23519715cebaf2b388e3e6969516","abstract_canon_sha256":"0931e54c733aaa0888477f3dceda2b4af8d2a6c0fceebbe038041053b171481b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:30:56.259838Z","signature_b64":"2hgrbtvBun2igh6K83UIPDfnKrliZ5Y4fx0Qp7kC7NC/FFMj0tnjOggyt3/hRcdqg7IcyMFMtiuwrZxwvbHlCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cde7d92c610b1548381fd9f15fc4b43da164e8fa20b1ae17a2f0ee85e52294d","last_reissued_at":"2026-07-05T10:30:56.258934Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:30:56.258934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DetectiveQA: Evaluating Long-Context Reasoning on Detective Novels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiasheng Ye, Linlin Li, Qipeng Guo, Qun Liu, Tianxiang Sun, Xiangyang Liu, Xiaoran Liu, Xipeng Qiu, Xuanjing Huang, Zhe Xu, Zhigeng Liu","submitted_at":"2024-09-04T06:28:22Z","abstract_excerpt":"Recently, significant efforts have been devoted to enhancing the long-context capabilities of Large Language Models (LLMs), particularly in long-context reasoning. To facilitate this research, we propose \\textbf{DetectiveQA}, a dataset specifically designed for narrative reasoning within long contexts. We leverage detective novels, averaging over 100k tokens, to create a dataset containing 1200 human-annotated questions in both Chinese and English, each paired with corresponding reference reasoning steps. Furthermore, we introduce a step-wise reasoning metric, which enhances the evaluation of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.02465","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/2409.02465/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2409.02465","created_at":"2026-07-05T10:30:56.259047+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.02465v2","created_at":"2026-07-05T10:30:56.259047+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.02465","created_at":"2026-07-05T10:30:56.259047+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTPH3EWGCCYV","created_at":"2026-07-05T10:30:56.259047+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTPH3EWGCCYVJA4B","created_at":"2026-07-05T10:30:56.259047+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTPH3EWG","created_at":"2026-07-05T10:30:56.259047+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10921","citing_title":"Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13511","citing_title":"Many-Shot CoT-ICL: Making In-Context Learning Truly Learn","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2507.05257","citing_title":"Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08016","citing_title":"Wiring the 'Why': A Unified Taxonomy and Survey of Abductive Reasoning in LLMs","ref_index":106,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP","json":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP.json","graph_json":"https://pith.science/api/pith-number/FTPH3EWGCCYVJA4B7WPRL7CLIP/graph.json","events_json":"https://pith.science/api/pith-number/FTPH3EWGCCYVJA4B7WPRL7CLIP/events.json","paper":"https://pith.science/paper/FTPH3EWG"},"agent_actions":{"view_html":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP","download_json":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP.json","view_paper":"https://pith.science/paper/FTPH3EWG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.02465&json=true","fetch_graph":"https://pith.science/api/pith-number/FTPH3EWGCCYVJA4B7WPRL7CLIP/graph.json","fetch_events":"https://pith.science/api/pith-number/FTPH3EWGCCYVJA4B7WPRL7CLIP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP/action/storage_attestation","attest_author":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP/action/author_attestation","sign_citation":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP/action/citation_signature","submit_replication":"https://pith.science/pith/FTPH3EWGCCYVJA4B7WPRL7CLIP/action/replication_record"}},"created_at":"2026-07-05T10:30:56.259047+00:00","updated_at":"2026-07-05T10:30:56.259047+00:00"}