{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UAFPP2553SVHFEUO3YHSZCI24P","short_pith_number":"pith:UAFPP255","schema_version":"1.0","canonical_sha256":"a00af7ebbddcaa72928ede0f2c891ae3dc029b982a77ab2ffb3f2fc27f86cfb2","source":{"kind":"arxiv","id":"2507.00163","version":2},"attestation_state":"computed","paper":{"title":"Prompting as Scientific Inquiry","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ari Holtzman, Chenhao Tan","submitted_at":"2025-06-30T18:11:25Z","abstract_excerpt":"Prompting is the primary method by which we study and control large language models. It is also one of the most powerful: nearly every major capability attributed to LLMs-few-shot learning, chain-of-thought, constitutional AI-was first unlocked through prompting. Yet prompting is rarely treated as science and is frequently frowned upon as alchemy. We argue that this is a category error. If we treat LLMs as a new kind of complex and opaque organism that is trained rather than programmed, then prompting is not a workaround: it is behavioral science. Mechanistic interpretability peers into the ne"},"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":"2507.00163","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-30T18:11:25Z","cross_cats_sorted":[],"title_canon_sha256":"9ce29786a9121f134aec446ffc4059d45bd511652ff47d5ec0d9d17e736e94b2","abstract_canon_sha256":"0e52b6fd64cc40f3570fd77b87fb04a626ea480656013ef8a81f1215f1888f3e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:47.088356Z","signature_b64":"54gDAJoh2igPh7Hq2s/5oz6dUndwCrSpRcTZP0HtgA5bz+Is4pdROe3wMK8kBYH2w8jgKJGizZl9TWXIDZcFBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a00af7ebbddcaa72928ede0f2c891ae3dc029b982a77ab2ffb3f2fc27f86cfb2","last_reissued_at":"2026-07-05T11:31:47.087846Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:47.087846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompting as Scientific Inquiry","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ari Holtzman, Chenhao Tan","submitted_at":"2025-06-30T18:11:25Z","abstract_excerpt":"Prompting is the primary method by which we study and control large language models. It is also one of the most powerful: nearly every major capability attributed to LLMs-few-shot learning, chain-of-thought, constitutional AI-was first unlocked through prompting. Yet prompting is rarely treated as science and is frequently frowned upon as alchemy. We argue that this is a category error. If we treat LLMs as a new kind of complex and opaque organism that is trained rather than programmed, then prompting is not a workaround: it is behavioral science. Mechanistic interpretability peers into the ne"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.00163","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/2507.00163/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":"2507.00163","created_at":"2026-07-05T11:31:47.087908+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.00163v2","created_at":"2026-07-05T11:31:47.087908+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.00163","created_at":"2026-07-05T11:31:47.087908+00:00"},{"alias_kind":"pith_short_12","alias_value":"UAFPP2553SVH","created_at":"2026-07-05T11:31:47.087908+00:00"},{"alias_kind":"pith_short_16","alias_value":"UAFPP2553SVHFEUO","created_at":"2026-07-05T11:31:47.087908+00:00"},{"alias_kind":"pith_short_8","alias_value":"UAFPP255","created_at":"2026-07-05T11:31:47.087908+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07479","citing_title":"Supervision versus Demonstration-Based In-Context Learning for Multiword Expression Classification","ref_index":180,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P","json":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P.json","graph_json":"https://pith.science/api/pith-number/UAFPP2553SVHFEUO3YHSZCI24P/graph.json","events_json":"https://pith.science/api/pith-number/UAFPP2553SVHFEUO3YHSZCI24P/events.json","paper":"https://pith.science/paper/UAFPP255"},"agent_actions":{"view_html":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P","download_json":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P.json","view_paper":"https://pith.science/paper/UAFPP255","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.00163&json=true","fetch_graph":"https://pith.science/api/pith-number/UAFPP2553SVHFEUO3YHSZCI24P/graph.json","fetch_events":"https://pith.science/api/pith-number/UAFPP2553SVHFEUO3YHSZCI24P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P/action/storage_attestation","attest_author":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P/action/author_attestation","sign_citation":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P/action/citation_signature","submit_replication":"https://pith.science/pith/UAFPP2553SVHFEUO3YHSZCI24P/action/replication_record"}},"created_at":"2026-07-05T11:31:47.087908+00:00","updated_at":"2026-07-05T11:31:47.087908+00:00"}