{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NXGCTTBAAOGPXR6BW5B4CYMXOZ","short_pith_number":"pith:NXGCTTBA","schema_version":"1.0","canonical_sha256":"6dcc29cc20038cfbc7c1b743c1619776541995a218ad6efbce9c25002e3d76bf","source":{"kind":"arxiv","id":"2404.00614","version":2},"attestation_state":"computed","paper":{"title":"Learning to Plan for Language Modeling from Unlabeled Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Florian Mai, Marie-Francine Moens, Nathan Cornille","submitted_at":"2024-03-31T09:04:01Z","abstract_excerpt":"By training to predict the next token in an unlabeled corpus, large language models learn to perform many tasks without any labeled data. However, their next-token-prediction objective arguably limits their performance in scenarios that require planning, such as writing a coherent article. In this paper, we train a module for planning the future writing process via a self-supervised learning objective. Given the textual context, this planning module learns to predict future abstract writing actions, which correspond to centroids in a clustered text embedding space. By conditioning on these act"},"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":"2404.00614","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-31T09:04:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2041331d292010f78ff9acd5909b2c62fab133eee13ed4470bbed871ab903026","abstract_canon_sha256":"453a44cecf35af996fe774af2eafa46a9265ef3af3e4ca61edb7076e8cc5055c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:28.509402Z","signature_b64":"qnM8zqao55RwhkftXxOaFkaQJnFrcAkJPKFeHwBGZC1yAHDQfkm9XRakjkljD+iVGYfipLNVPsbKiuBacZn+Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6dcc29cc20038cfbc7c1b743c1619776541995a218ad6efbce9c25002e3d76bf","last_reissued_at":"2026-07-05T08:50:28.508951Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:28.508951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Plan for Language Modeling from Unlabeled Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Florian Mai, Marie-Francine Moens, Nathan Cornille","submitted_at":"2024-03-31T09:04:01Z","abstract_excerpt":"By training to predict the next token in an unlabeled corpus, large language models learn to perform many tasks without any labeled data. However, their next-token-prediction objective arguably limits their performance in scenarios that require planning, such as writing a coherent article. In this paper, we train a module for planning the future writing process via a self-supervised learning objective. Given the textual context, this planning module learns to predict future abstract writing actions, which correspond to centroids in a clustered text embedding space. By conditioning on these act"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.00614","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/2404.00614/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":"2404.00614","created_at":"2026-07-05T08:50:28.509023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.00614v2","created_at":"2026-07-05T08:50:28.509023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.00614","created_at":"2026-07-05T08:50:28.509023+00:00"},{"alias_kind":"pith_short_12","alias_value":"NXGCTTBAAOGP","created_at":"2026-07-05T08:50:28.509023+00:00"},{"alias_kind":"pith_short_16","alias_value":"NXGCTTBAAOGPXR6B","created_at":"2026-07-05T08:50:28.509023+00:00"},{"alias_kind":"pith_short_8","alias_value":"NXGCTTBA","created_at":"2026-07-05T08:50:28.509023+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03624","citing_title":"LoopMTP: A looped transformer guided by latent multi-token prediction","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ","json":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ.json","graph_json":"https://pith.science/api/pith-number/NXGCTTBAAOGPXR6BW5B4CYMXOZ/graph.json","events_json":"https://pith.science/api/pith-number/NXGCTTBAAOGPXR6BW5B4CYMXOZ/events.json","paper":"https://pith.science/paper/NXGCTTBA"},"agent_actions":{"view_html":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ","download_json":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ.json","view_paper":"https://pith.science/paper/NXGCTTBA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.00614&json=true","fetch_graph":"https://pith.science/api/pith-number/NXGCTTBAAOGPXR6BW5B4CYMXOZ/graph.json","fetch_events":"https://pith.science/api/pith-number/NXGCTTBAAOGPXR6BW5B4CYMXOZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ/action/storage_attestation","attest_author":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ/action/author_attestation","sign_citation":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ/action/citation_signature","submit_replication":"https://pith.science/pith/NXGCTTBAAOGPXR6BW5B4CYMXOZ/action/replication_record"}},"created_at":"2026-07-05T08:50:28.509023+00:00","updated_at":"2026-07-05T08:50:28.509023+00:00"}