{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:COUQ6OIQ2YDCWRXVHSUW35EZKS","short_pith_number":"pith:COUQ6OIQ","schema_version":"1.0","canonical_sha256":"13a90f3910d6062b46f53ca96df49954b8fd9a643c7b6c3a10203f560440d1fd","source":{"kind":"arxiv","id":"2602.00906","version":7},"attestation_state":"computed","paper":{"title":"Hallucination is a Consequence of Space-Optimality: A Rate-Distortion Theorem for Membership Testing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Even with perfect data and training, limited capacity forces LLMs to assign high confidence to some non-facts","cross_cats":["cs.AI","cs.CL","cs.DS","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Anxin Guo, Jingwei Li","submitted_at":"2026-01-31T21:18:28Z","abstract_excerpt":"Large language models often hallucinate with high confidence on \"random facts\" that lack inferable patterns. We formalize the memorization of such facts as a membership testing problem, unifying the discrete error metrics of Bloom filters with the continuous log-loss of LLMs. By analyzing this problem in the regime where facts are sparse in the universe of plausible claims, we establish a rate-distortion theorem: the optimal memory efficiency is characterized by the minimum KL divergence between score distributions on facts and non-facts. This theoretical framework provides a distinctive expla"},"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":true},"canonical_record":{"source":{"id":"2602.00906","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-31T21:18:28Z","cross_cats_sorted":["cs.AI","cs.CL","cs.DS","cs.IT","math.IT"],"title_canon_sha256":"09c1a2f0b896d2323af97c5dfd4a990494b8d768f3a2d1265f5a5f067db8c74b","abstract_canon_sha256":"ded0d6a9ce377f077ab0fd83b96766f8267820895964baf8db5a43b51069eaf6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:41.728035Z","signature_b64":"TIh+AGQvzPmAjOZV7Jgi8PzphEPCqqRherNOjfz0IlX+mPoT9FbJD9oO9ozuFj5FsX3qEkZAxjzV1/ZK1+0MBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"13a90f3910d6062b46f53ca96df49954b8fd9a643c7b6c3a10203f560440d1fd","last_reissued_at":"2026-06-02T01:03:41.727527Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:41.727527Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hallucination is a Consequence of Space-Optimality: A Rate-Distortion Theorem for Membership Testing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Even with perfect data and training, limited capacity forces LLMs to assign high confidence to some non-facts","cross_cats":["cs.AI","cs.CL","cs.DS","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Anxin Guo, Jingwei Li","submitted_at":"2026-01-31T21:18:28Z","abstract_excerpt":"Large language models often hallucinate with high confidence on \"random facts\" that lack inferable patterns. We formalize the memorization of such facts as a membership testing problem, unifying the discrete error metrics of Bloom filters with the continuous log-loss of LLMs. By analyzing this problem in the regime where facts are sparse in the universe of plausible claims, we establish a rate-distortion theorem: the optimal memory efficiency is characterized by the minimum KL divergence between score distributions on facts and non-facts. This theoretical framework provides a distinctive expla"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"even with optimal training, perfect data, and a simplified closed world setting, the information-theoretically optimal strategy under limited capacity is not to abstain or forget, but to assign high confidence to some non-facts, resulting in hallucination.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The regime in which facts are sparse in the universe of plausible claims, together with the modeling choice that unifies discrete Bloom-filter error with continuous log-loss under a single rate-distortion objective.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Hallucinations are the space-optimal behavior for limited-capacity models performing membership testing on sparse facts, as shown by a rate-distortion theorem that equates optimal memory use to minimum KL divergence between fact and non-fact score distributions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Even with perfect data and training, limited capacity forces LLMs to assign high confidence to some non-facts","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"38afcfcfca17f0f53a74524565a3b575ad1a7c2b5c079ea23961b78a198b077c"},"source":{"id":"2602.00906","kind":"arxiv","version":7},"verdict":{"id":"d95c20da-8cad-440e-93ee-c14efa3fcfd6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T08:33:35.435664Z","strongest_claim":"even with optimal training, perfect data, and a simplified closed world setting, the information-theoretically optimal strategy under limited capacity is not to abstain or forget, but to assign high confidence to some non-facts, resulting in hallucination.","one_line_summary":"Hallucinations are the space-optimal behavior for limited-capacity models performing membership testing on sparse facts, as shown by a rate-distortion theorem that equates optimal memory use to minimum KL divergence between fact and non-fact score distributions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The regime in which facts are sparse in the universe of plausible claims, together with the modeling choice that unifies discrete Bloom-filter error with continuous log-loss under a single rate-distortion objective.","pith_extraction_headline":"Even with perfect data and training, limited capacity forces LLMs to assign high confidence to some non-facts"},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.00906/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":2,"snapshot_sha256":"076041c912f00d4f04be6d3068f4d74b681878e687840297e73be32065210925"},"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":"2602.00906","created_at":"2026-06-02T01:03:41.727591+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.00906v7","created_at":"2026-06-02T01:03:41.727591+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.00906","created_at":"2026-06-02T01:03:41.727591+00:00"},{"alias_kind":"pith_short_12","alias_value":"COUQ6OIQ2YDC","created_at":"2026-06-02T01:03:41.727591+00:00"},{"alias_kind":"pith_short_16","alias_value":"COUQ6OIQ2YDCWRXV","created_at":"2026-06-02T01:03:41.727591+00:00"},{"alias_kind":"pith_short_8","alias_value":"COUQ6OIQ","created_at":"2026-06-02T01:03:41.727591+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS","json":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS.json","graph_json":"https://pith.science/api/pith-number/COUQ6OIQ2YDCWRXVHSUW35EZKS/graph.json","events_json":"https://pith.science/api/pith-number/COUQ6OIQ2YDCWRXVHSUW35EZKS/events.json","paper":"https://pith.science/paper/COUQ6OIQ"},"agent_actions":{"view_html":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS","download_json":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS.json","view_paper":"https://pith.science/paper/COUQ6OIQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.00906&json=true","fetch_graph":"https://pith.science/api/pith-number/COUQ6OIQ2YDCWRXVHSUW35EZKS/graph.json","fetch_events":"https://pith.science/api/pith-number/COUQ6OIQ2YDCWRXVHSUW35EZKS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS/action/storage_attestation","attest_author":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS/action/author_attestation","sign_citation":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS/action/citation_signature","submit_replication":"https://pith.science/pith/COUQ6OIQ2YDCWRXVHSUW35EZKS/action/replication_record"}},"created_at":"2026-06-02T01:03:41.727591+00:00","updated_at":"2026-06-02T01:03:41.727591+00:00"}