{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PIWGKQ3535OML54JX3MWJXQX4W","short_pith_number":"pith:PIWGKQ35","schema_version":"1.0","canonical_sha256":"7a2c65437ddf5cc5f789bed964de17e59d8af877dc3b782aaf5e6254445191ab","source":{"kind":"arxiv","id":"2304.08460","version":3},"attestation_state":"computed","paper":{"title":"LongForm: Effective Instruction Tuning with Reverse Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Abdullatif K\\\"oksal, Anna Korhonen, Hinrich Sch\\\"utze, Timo Schick","submitted_at":"2023-04-17T17:36:35Z","abstract_excerpt":"Instruction tuning enables language models to more effectively generalize and better follow user intent. However, obtaining instruction data is costly and challenging. Prior work employs methods such as expensive human annotation, crowd-sourced datasets with alignment issues, and generating noisy examples via LLMs. We introduce the LongForm-C dataset, which is created by reverse instructions. We generate instructions via LLMs for human-written corpus examples using reverse instructions. First we select a diverse set of human-written documents from corpora such as C4 and Wikipedia; then we gene"},"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":"2304.08460","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-17T17:36:35Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0bc2cde66b55a6b8045394d903b5e609955957d0be22f34687755e1b1260aa9d","abstract_canon_sha256":"24e08c3c2ef3fda35ad6cbaaee115e9cec95d44def6efa646cc82d583d11a0fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:05.881267Z","signature_b64":"nMRHXrgZ9D6QYdRrrBZgXyc4IE0x63vYbYY+6PqV2xa5sAqovZ/FsIjAMi+DmRzUZClg1Vgc+XiZ9vEafeJEAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a2c65437ddf5cc5f789bed964de17e59d8af877dc3b782aaf5e6254445191ab","last_reissued_at":"2026-07-05T09:15:05.880803Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:05.880803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LongForm: Effective Instruction Tuning with Reverse Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Abdullatif K\\\"oksal, Anna Korhonen, Hinrich Sch\\\"utze, Timo Schick","submitted_at":"2023-04-17T17:36:35Z","abstract_excerpt":"Instruction tuning enables language models to more effectively generalize and better follow user intent. However, obtaining instruction data is costly and challenging. Prior work employs methods such as expensive human annotation, crowd-sourced datasets with alignment issues, and generating noisy examples via LLMs. We introduce the LongForm-C dataset, which is created by reverse instructions. We generate instructions via LLMs for human-written corpus examples using reverse instructions. First we select a diverse set of human-written documents from corpora such as C4 and Wikipedia; then we gene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.08460","kind":"arxiv","version":3},"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/2304.08460/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":"2304.08460","created_at":"2026-07-05T09:15:05.880858+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.08460v3","created_at":"2026-07-05T09:15:05.880858+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.08460","created_at":"2026-07-05T09:15:05.880858+00:00"},{"alias_kind":"pith_short_12","alias_value":"PIWGKQ3535OM","created_at":"2026-07-05T09:15:05.880858+00:00"},{"alias_kind":"pith_short_16","alias_value":"PIWGKQ3535OML54J","created_at":"2026-07-05T09:15:05.880858+00:00"},{"alias_kind":"pith_short_8","alias_value":"PIWGKQ35","created_at":"2026-07-05T09:15:05.880858+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06145","citing_title":"Prompting Complexity: Shortest Prompts for Texts and Behaviors in LLMs","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2401.10020","citing_title":"Self-Rewarding Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2305.14314","citing_title":"QLoRA: Efficient Finetuning of Quantized LLMs","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W","json":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W.json","graph_json":"https://pith.science/api/pith-number/PIWGKQ3535OML54JX3MWJXQX4W/graph.json","events_json":"https://pith.science/api/pith-number/PIWGKQ3535OML54JX3MWJXQX4W/events.json","paper":"https://pith.science/paper/PIWGKQ35"},"agent_actions":{"view_html":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W","download_json":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W.json","view_paper":"https://pith.science/paper/PIWGKQ35","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.08460&json=true","fetch_graph":"https://pith.science/api/pith-number/PIWGKQ3535OML54JX3MWJXQX4W/graph.json","fetch_events":"https://pith.science/api/pith-number/PIWGKQ3535OML54JX3MWJXQX4W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W/action/storage_attestation","attest_author":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W/action/author_attestation","sign_citation":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W/action/citation_signature","submit_replication":"https://pith.science/pith/PIWGKQ3535OML54JX3MWJXQX4W/action/replication_record"}},"created_at":"2026-07-05T09:15:05.880858+00:00","updated_at":"2026-07-05T09:15:05.880858+00:00"}