{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HMXDVGXDEKUTWORIMIJJFDGXUB","short_pith_number":"pith:HMXDVGXD","schema_version":"1.0","canonical_sha256":"3b2e3a9ae322a93b3a286212928cd7a047c6364a91826626006176f139c12d61","source":{"kind":"arxiv","id":"2608.07208","version":1},"attestation_state":"computed","paper":{"title":"Measuring Concept Content in Text from LLM Activations: ESG Evidence from Concept Vectors and Linear Probes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","econ.GN","q-fin.EC"],"primary_cat":"cs.CL","authors_text":"Amirhossein Zohrehvand, Luc Hazenoot, Zhaochun Ren","submitted_at":"2026-08-07T13:21:50Z","abstract_excerpt":"Existing measures of how much a text is about a concept read the surface of the text: dictionary word shares, topic proportions, embedding similarities. They score the words a text uses, not the judgment a reader forms about it. Recent work has shown that a gap exists in what Large Language Models (LLMs) know internally versus what they express in their response. This paper asks whether that internal knowledge, read by monitoring the activations of frozen, out-of-the-box LLMs, can stand in for task-specific fine-tuning when measuring concept content, and which extraction method reads it best. "},"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":"2608.07208","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-07T13:21:50Z","cross_cats_sorted":["cs.AI","cs.LG","econ.GN","q-fin.EC"],"title_canon_sha256":"c5122e2dedde73ccee75791a12195da35f58d7e606316b69e60ffed4b41d162a","abstract_canon_sha256":"eae93556e066b77ff8a8c6d8f8615f8fc5207537d0cae3b90abfc0c6c28392c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:14:38.412674Z","signature_b64":"rxQ4UY+3REeaNuJ1Fen4/pkyFHjfcU1s306wL1CMJ5DhHzurInvZLAPa2kWQ8zSGyldJTrzEDOCwg8VlBh44Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b2e3a9ae322a93b3a286212928cd7a047c6364a91826626006176f139c12d61","last_reissued_at":"2026-08-10T01:14:38.409895Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:14:38.409895Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Measuring Concept Content in Text from LLM Activations: ESG Evidence from Concept Vectors and Linear Probes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","econ.GN","q-fin.EC"],"primary_cat":"cs.CL","authors_text":"Amirhossein Zohrehvand, Luc Hazenoot, Zhaochun Ren","submitted_at":"2026-08-07T13:21:50Z","abstract_excerpt":"Existing measures of how much a text is about a concept read the surface of the text: dictionary word shares, topic proportions, embedding similarities. They score the words a text uses, not the judgment a reader forms about it. Recent work has shown that a gap exists in what Large Language Models (LLMs) know internally versus what they express in their response. This paper asks whether that internal knowledge, read by monitoring the activations of frozen, out-of-the-box LLMs, can stand in for task-specific fine-tuning when measuring concept content, and which extraction method reads it best. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07208","kind":"arxiv","version":1},"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/2608.07208/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":"2608.07208","created_at":"2026-08-10T01:14:38.411127+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.07208v1","created_at":"2026-08-10T01:14:38.411127+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07208","created_at":"2026-08-10T01:14:38.411127+00:00"},{"alias_kind":"pith_short_12","alias_value":"HMXDVGXDEKUT","created_at":"2026-08-10T01:14:38.411127+00:00"},{"alias_kind":"pith_short_16","alias_value":"HMXDVGXDEKUTWORI","created_at":"2026-08-10T01:14:38.411127+00:00"},{"alias_kind":"pith_short_8","alias_value":"HMXDVGXD","created_at":"2026-08-10T01:14:38.411127+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/HMXDVGXDEKUTWORIMIJJFDGXUB","json":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB.json","graph_json":"https://pith.science/api/pith-number/HMXDVGXDEKUTWORIMIJJFDGXUB/graph.json","events_json":"https://pith.science/api/pith-number/HMXDVGXDEKUTWORIMIJJFDGXUB/events.json","paper":"https://pith.science/paper/HMXDVGXD"},"agent_actions":{"view_html":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB","download_json":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB.json","view_paper":"https://pith.science/paper/HMXDVGXD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.07208&json=true","fetch_graph":"https://pith.science/api/pith-number/HMXDVGXDEKUTWORIMIJJFDGXUB/graph.json","fetch_events":"https://pith.science/api/pith-number/HMXDVGXDEKUTWORIMIJJFDGXUB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB/action/storage_attestation","attest_author":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB/action/author_attestation","sign_citation":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB/action/citation_signature","submit_replication":"https://pith.science/pith/HMXDVGXDEKUTWORIMIJJFDGXUB/action/replication_record"}},"created_at":"2026-08-10T01:14:38.411127+00:00","updated_at":"2026-08-10T01:14:38.411127+00:00"}