{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GBTCRHS6VNPHDH233SCBSGHYW3","short_pith_number":"pith:GBTCRHS6","schema_version":"1.0","canonical_sha256":"3066289e5eab5e719f5bdc841918f8b6f23accfe59b7e80fcaf2e31ec9355f4a","source":{"kind":"arxiv","id":"2505.17071","version":1},"attestation_state":"computed","paper":{"title":"What's in a prompt? Language models encode literary style in prompt embeddings","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christopher Earls, Haley Moller, Nicolas Boull\\'e, Rapha\\\"el Sarfati, Toni J. B. Liu","submitted_at":"2025-05-19T15:56:13Z","abstract_excerpt":"Large language models use high-dimensional latent spaces to encode and process textual information. Much work has investigated how the conceptual content of words translates into geometrical relationships between their vector representations. Fewer studies analyze how the cumulative information of an entire prompt becomes condensed into individual embeddings under the action of transformer layers. We use literary pieces to show that information about intangible, rather than factual, aspects of the prompt are contained in deep representations. We observe that short excerpts (10 - 100 tokens) fr"},"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":"2505.17071","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-19T15:56:13Z","cross_cats_sorted":[],"title_canon_sha256":"6c6972db350a6fb7b6809a67f43e49a694487fdb6b515d26a47cbafef6ec8a0c","abstract_canon_sha256":"53fd857e40ee4a7d1c2e75e74887e01794727a4d3a4ac888a52c94822b688568"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:46.610012Z","signature_b64":"d4grtX+rfuUFL0HLuEbe/9JD+iVP99bEJIs5Ku4Q34kVpjdtgFe8UyYQi47kFFspQwWqS6RFzLwqqfZtaG5jDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3066289e5eab5e719f5bdc841918f8b6f23accfe59b7e80fcaf2e31ec9355f4a","last_reissued_at":"2026-07-05T11:07:46.609494Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:46.609494Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What's in a prompt? Language models encode literary style in prompt embeddings","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christopher Earls, Haley Moller, Nicolas Boull\\'e, Rapha\\\"el Sarfati, Toni J. B. Liu","submitted_at":"2025-05-19T15:56:13Z","abstract_excerpt":"Large language models use high-dimensional latent spaces to encode and process textual information. Much work has investigated how the conceptual content of words translates into geometrical relationships between their vector representations. Fewer studies analyze how the cumulative information of an entire prompt becomes condensed into individual embeddings under the action of transformer layers. We use literary pieces to show that information about intangible, rather than factual, aspects of the prompt are contained in deep representations. We observe that short excerpts (10 - 100 tokens) fr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17071","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/2505.17071/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":"2505.17071","created_at":"2026-07-05T11:07:46.609556+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.17071v1","created_at":"2026-07-05T11:07:46.609556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17071","created_at":"2026-07-05T11:07:46.609556+00:00"},{"alias_kind":"pith_short_12","alias_value":"GBTCRHS6VNPH","created_at":"2026-07-05T11:07:46.609556+00:00"},{"alias_kind":"pith_short_16","alias_value":"GBTCRHS6VNPHDH23","created_at":"2026-07-05T11:07:46.609556+00:00"},{"alias_kind":"pith_short_8","alias_value":"GBTCRHS6","created_at":"2026-07-05T11:07:46.609556+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/GBTCRHS6VNPHDH233SCBSGHYW3","json":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3.json","graph_json":"https://pith.science/api/pith-number/GBTCRHS6VNPHDH233SCBSGHYW3/graph.json","events_json":"https://pith.science/api/pith-number/GBTCRHS6VNPHDH233SCBSGHYW3/events.json","paper":"https://pith.science/paper/GBTCRHS6"},"agent_actions":{"view_html":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3","download_json":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3.json","view_paper":"https://pith.science/paper/GBTCRHS6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.17071&json=true","fetch_graph":"https://pith.science/api/pith-number/GBTCRHS6VNPHDH233SCBSGHYW3/graph.json","fetch_events":"https://pith.science/api/pith-number/GBTCRHS6VNPHDH233SCBSGHYW3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3/action/storage_attestation","attest_author":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3/action/author_attestation","sign_citation":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3/action/citation_signature","submit_replication":"https://pith.science/pith/GBTCRHS6VNPHDH233SCBSGHYW3/action/replication_record"}},"created_at":"2026-07-05T11:07:46.609556+00:00","updated_at":"2026-07-05T11:07:46.609556+00:00"}