{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:R33XLFFBMQ4WYGZPQOTKNRKWU5","short_pith_number":"pith:R33XLFFB","schema_version":"1.0","canonical_sha256":"8ef77594a164396c1b2f83a6a6c556a7453a7320d2cd6edc1e4f5644c285373c","source":{"kind":"arxiv","id":"2409.14254","version":1},"attestation_state":"computed","paper":{"title":"Instruction Following without Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christopher D. Manning, John Hewitt, Nelson F. Liu, Percy Liang","submitted_at":"2024-09-21T22:36:22Z","abstract_excerpt":"Instruction tuning commonly means finetuning a language model on instruction-response pairs. We discover two forms of adaptation (tuning) that are deficient compared to instruction tuning, yet still yield instruction following; we call this implicit instruction tuning. We first find that instruction-response pairs are not necessary: training solely on responses, without any corresponding instructions, yields instruction following. This suggests pretrained models have an instruction-response mapping which is revealed by teaching the model the desired distribution of responses. However, we then "},"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":"2409.14254","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-21T22:36:22Z","cross_cats_sorted":[],"title_canon_sha256":"1d8b99b0d373e941ddb58fde1ab3a41db49e53dbbc9c9e964b80dce9d3228bc2","abstract_canon_sha256":"ddd42d0f343e820d5c27a1d50be653d0c524f63ed93dfeb80d695a504725d848"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:13.871344Z","signature_b64":"DgM7S27mwx695Trm7mBqaI0b4C0NJEp85kwwmsN/BRH6eG96eklR6vBYo7FCNCidw28fu3ClFwCM+Zq0iyrICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ef77594a164396c1b2f83a6a6c556a7453a7320d2cd6edc1e4f5644c285373c","last_reissued_at":"2026-07-05T09:10:13.870922Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:13.870922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Instruction Following without Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christopher D. Manning, John Hewitt, Nelson F. Liu, Percy Liang","submitted_at":"2024-09-21T22:36:22Z","abstract_excerpt":"Instruction tuning commonly means finetuning a language model on instruction-response pairs. We discover two forms of adaptation (tuning) that are deficient compared to instruction tuning, yet still yield instruction following; we call this implicit instruction tuning. We first find that instruction-response pairs are not necessary: training solely on responses, without any corresponding instructions, yields instruction following. This suggests pretrained models have an instruction-response mapping which is revealed by teaching the model the desired distribution of responses. However, we then "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.14254","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/2409.14254/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":"2409.14254","created_at":"2026-07-05T09:10:13.870978+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.14254v1","created_at":"2026-07-05T09:10:13.870978+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.14254","created_at":"2026-07-05T09:10:13.870978+00:00"},{"alias_kind":"pith_short_12","alias_value":"R33XLFFBMQ4W","created_at":"2026-07-05T09:10:13.870978+00:00"},{"alias_kind":"pith_short_16","alias_value":"R33XLFFBMQ4WYGZP","created_at":"2026-07-05T09:10:13.870978+00:00"},{"alias_kind":"pith_short_8","alias_value":"R33XLFFB","created_at":"2026-07-05T09:10:13.870978+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.22598","citing_title":"RExBench: Can coding agents autonomously implement AI research extensions?","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5","json":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5.json","graph_json":"https://pith.science/api/pith-number/R33XLFFBMQ4WYGZPQOTKNRKWU5/graph.json","events_json":"https://pith.science/api/pith-number/R33XLFFBMQ4WYGZPQOTKNRKWU5/events.json","paper":"https://pith.science/paper/R33XLFFB"},"agent_actions":{"view_html":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5","download_json":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5.json","view_paper":"https://pith.science/paper/R33XLFFB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.14254&json=true","fetch_graph":"https://pith.science/api/pith-number/R33XLFFBMQ4WYGZPQOTKNRKWU5/graph.json","fetch_events":"https://pith.science/api/pith-number/R33XLFFBMQ4WYGZPQOTKNRKWU5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5/action/storage_attestation","attest_author":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5/action/author_attestation","sign_citation":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5/action/citation_signature","submit_replication":"https://pith.science/pith/R33XLFFBMQ4WYGZPQOTKNRKWU5/action/replication_record"}},"created_at":"2026-07-05T09:10:13.870978+00:00","updated_at":"2026-07-05T09:10:13.870978+00:00"}