{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4MHSVWTAYPQJ33LWNJ7N2ATELJ","short_pith_number":"pith:4MHSVWTA","schema_version":"1.0","canonical_sha256":"e30f2ada60c3e09ded766a7edd02645a443b09728a65fefb5cdcca938a9004e9","source":{"kind":"arxiv","id":"2601.02023","version":2},"attestation_state":"computed","paper":{"title":"Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Retrieval, Reasoning, and Hallucination in Long-Context LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amirali Ebrahimzadeh, Seyyed M. Salili","submitted_at":"2026-01-05T11:30:56Z","abstract_excerpt":"As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distributed in real-world corpora. We investigate how fact placement, corpus-level distributions, and anti-hallucination (\"Don't Make It Up\") prompts influence model behavior by introducing a model-agnostic extended needle-in-a-haystack benchmark designed for scalability, which we apply to evaluate Gemini-2.5-flash, ChatGPT-5-mini, Claude-4.5-haiku, and Deepseek-v3.2-chat. Unlike prior work, we separately"},"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":"2601.02023","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-05T11:30:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"dfdad30e7e18eda9b84a257342d83158cb2c53417d0fe17592e26220f1280bd4","abstract_canon_sha256":"3210e0d7c00fb24f1dd8ae51df558810e7a1bfe5f8cb51ca5dac23ec55a5091a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-16T00:21:44.328702Z","signature_b64":"Dsc/F6irQvDRVOPCl1Rg37XcZcTYigW++dc0JCn610SVSRzIeuTcErS4W8Er+3Q9pmQhRqSbh14kVg59JpG6AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e30f2ada60c3e09ded766a7edd02645a443b09728a65fefb5cdcca938a9004e9","last_reissued_at":"2026-07-16T00:21:44.327890Z","signature_status":"signed_v1","first_computed_at":"2026-07-16T00:21:44.327890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Retrieval, Reasoning, and Hallucination in Long-Context LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amirali Ebrahimzadeh, Seyyed M. Salili","submitted_at":"2026-01-05T11:30:56Z","abstract_excerpt":"As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distributed in real-world corpora. We investigate how fact placement, corpus-level distributions, and anti-hallucination (\"Don't Make It Up\") prompts influence model behavior by introducing a model-agnostic extended needle-in-a-haystack benchmark designed for scalability, which we apply to evaluate Gemini-2.5-flash, ChatGPT-5-mini, Claude-4.5-haiku, and Deepseek-v3.2-chat. Unlike prior work, we separately"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.02023","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/2601.02023/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":"2601.02023","created_at":"2026-07-16T00:21:44.328294+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.02023v2","created_at":"2026-07-16T00:21:44.328294+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.02023","created_at":"2026-07-16T00:21:44.328294+00:00"},{"alias_kind":"pith_short_12","alias_value":"4MHSVWTAYPQJ","created_at":"2026-07-16T00:21:44.328294+00:00"},{"alias_kind":"pith_short_16","alias_value":"4MHSVWTAYPQJ33LW","created_at":"2026-07-16T00:21:44.328294+00:00"},{"alias_kind":"pith_short_8","alias_value":"4MHSVWTA","created_at":"2026-07-16T00:21:44.328294+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ","json":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ.json","graph_json":"https://pith.science/api/pith-number/4MHSVWTAYPQJ33LWNJ7N2ATELJ/graph.json","events_json":"https://pith.science/api/pith-number/4MHSVWTAYPQJ33LWNJ7N2ATELJ/events.json","paper":"https://pith.science/paper/4MHSVWTA"},"agent_actions":{"view_html":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ","download_json":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ.json","view_paper":"https://pith.science/paper/4MHSVWTA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.02023&json=true","fetch_graph":"https://pith.science/api/pith-number/4MHSVWTAYPQJ33LWNJ7N2ATELJ/graph.json","fetch_events":"https://pith.science/api/pith-number/4MHSVWTAYPQJ33LWNJ7N2ATELJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ/action/storage_attestation","attest_author":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ/action/author_attestation","sign_citation":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ/action/citation_signature","submit_replication":"https://pith.science/pith/4MHSVWTAYPQJ33LWNJ7N2ATELJ/action/replication_record"}},"created_at":"2026-07-16T00:21:44.328294+00:00","updated_at":"2026-07-16T00:21:44.328294+00:00"}