{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PCSYYJBHGABZREFZIZCDGCD5SE","short_pith_number":"pith:PCSYYJBH","schema_version":"1.0","canonical_sha256":"78a58c242730039890b9464433087d913effc69d5ff304cef478f4443c8c17d5","source":{"kind":"arxiv","id":"2411.14257","version":2},"attestation_state":"computed","paper":{"title":"Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Javier Ferrando, Neel Nanda, Oscar Obeso, Senthooran Rajamanoharan","submitted_at":"2024-11-21T16:05:58Z","abstract_excerpt":"Hallucinations in large language models are a widespread problem, yet the mechanisms behind whether models will hallucinate are poorly understood, limiting our ability to solve this problem. Using sparse autoencoders as an interpretability tool, we discover that a key part of these mechanisms is entity recognition, where the model detects if an entity is one it can recall facts about. Sparse autoencoders uncover meaningful directions in the representation space, these detect whether the model recognizes an entity, e.g. detecting it doesn't know about an athlete or a movie. This suggests that m"},"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":"2411.14257","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T16:05:58Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5fdb1e59d109c10284f6e0f83fe3e0825252a1950e2a649a937d1ff865469268","abstract_canon_sha256":"38f68291cc3fb20943a4c95e3e89bef4c3d80ad135b24b002ca997f4a05481a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:11:28.939696Z","signature_b64":"PzdIw0AVb4jnj50dbfNYUTMbjD7gL2hCjG+sLBMy2NV3F18AXbxlT4ULDndZqFV+k47AwMs56bJoX3ssfsiHDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78a58c242730039890b9464433087d913effc69d5ff304cef478f4443c8c17d5","last_reissued_at":"2026-07-05T10:11:28.939203Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:11:28.939203Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Javier Ferrando, Neel Nanda, Oscar Obeso, Senthooran Rajamanoharan","submitted_at":"2024-11-21T16:05:58Z","abstract_excerpt":"Hallucinations in large language models are a widespread problem, yet the mechanisms behind whether models will hallucinate are poorly understood, limiting our ability to solve this problem. Using sparse autoencoders as an interpretability tool, we discover that a key part of these mechanisms is entity recognition, where the model detects if an entity is one it can recall facts about. Sparse autoencoders uncover meaningful directions in the representation space, these detect whether the model recognizes an entity, e.g. detecting it doesn't know about an athlete or a movie. This suggests that m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.14257","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/2411.14257/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":"2411.14257","created_at":"2026-07-05T10:11:28.939260+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.14257v2","created_at":"2026-07-05T10:11:28.939260+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.14257","created_at":"2026-07-05T10:11:28.939260+00:00"},{"alias_kind":"pith_short_12","alias_value":"PCSYYJBHGABZ","created_at":"2026-07-05T10:11:28.939260+00:00"},{"alias_kind":"pith_short_16","alias_value":"PCSYYJBHGABZREFZ","created_at":"2026-07-05T10:11:28.939260+00:00"},{"alias_kind":"pith_short_8","alias_value":"PCSYYJBH","created_at":"2026-07-05T10:11:28.939260+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07670","citing_title":"Does Bielik Know What It Doesn't Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21595","citing_title":"Per-Entity Bias Mapping for AI Visibility: Why Brand Mentions Require Entity-Specific Calibration","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17187","citing_title":"PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14192","citing_title":"Why Retrieval-Augmented Generation Fails: A Graph Perspective","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14270","citing_title":"Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19974","citing_title":"Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE","json":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE.json","graph_json":"https://pith.science/api/pith-number/PCSYYJBHGABZREFZIZCDGCD5SE/graph.json","events_json":"https://pith.science/api/pith-number/PCSYYJBHGABZREFZIZCDGCD5SE/events.json","paper":"https://pith.science/paper/PCSYYJBH"},"agent_actions":{"view_html":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE","download_json":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE.json","view_paper":"https://pith.science/paper/PCSYYJBH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.14257&json=true","fetch_graph":"https://pith.science/api/pith-number/PCSYYJBHGABZREFZIZCDGCD5SE/graph.json","fetch_events":"https://pith.science/api/pith-number/PCSYYJBHGABZREFZIZCDGCD5SE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE/action/storage_attestation","attest_author":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE/action/author_attestation","sign_citation":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE/action/citation_signature","submit_replication":"https://pith.science/pith/PCSYYJBHGABZREFZIZCDGCD5SE/action/replication_record"}},"created_at":"2026-07-05T10:11:28.939260+00:00","updated_at":"2026-07-05T10:11:28.939260+00:00"}