{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZKO4P5EBGEBZUX4VWNYQVZ7JNT","short_pith_number":"pith:ZKO4P5EB","schema_version":"1.0","canonical_sha256":"ca9dc7f48131039a5f95b3710ae7e96cd03af38a2ca3d448e75cacbf81367458","source":{"kind":"arxiv","id":"2310.05707","version":4},"attestation_state":"computed","paper":{"title":"Guiding Language Model Reasoning with Planning Tokens","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alessandro Sordoni, Lucas Caccia, Oleksiy Ostapenko, William Yang Wang, Xingdi Yuan, Xinyi Wang","submitted_at":"2023-10-09T13:29:37Z","abstract_excerpt":"Large language models (LLMs) have recently attracted considerable interest for their ability to perform complex reasoning tasks, such as chain-of-thought (CoT) reasoning. However, most of the existing approaches to enhance this ability rely heavily on data-driven methods, while neglecting the structural aspects of the model's reasoning capacity. To encourage a more structural generation of CoT steps, we propose a hierarchical generation scheme: we let the LM generate a planning token at the start of each reasoning step, intuitively serving as a high-level plan of the current step, and add thei"},"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":"2310.05707","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T13:29:37Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"ce61a295922b26b27d6f797ff068225470ebdc069acac4b483fbaf939397c2c6","abstract_canon_sha256":"82b7d9b531d7c32ac6e6b371d5c5b2f63e0d78a24819a3f04b74fab11821a3cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:52:51.212803Z","signature_b64":"Bs2z4LXQf3lbhVT8V8JrGwx4UWdrfLoXlGL5pA9wBPRSfZvm0AfxKZYMokiXtqtVmCLbXuV4guZlHtBa6u0PAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca9dc7f48131039a5f95b3710ae7e96cd03af38a2ca3d448e75cacbf81367458","last_reissued_at":"2026-07-05T08:52:51.212298Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:52:51.212298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Guiding Language Model Reasoning with Planning Tokens","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alessandro Sordoni, Lucas Caccia, Oleksiy Ostapenko, William Yang Wang, Xingdi Yuan, Xinyi Wang","submitted_at":"2023-10-09T13:29:37Z","abstract_excerpt":"Large language models (LLMs) have recently attracted considerable interest for their ability to perform complex reasoning tasks, such as chain-of-thought (CoT) reasoning. However, most of the existing approaches to enhance this ability rely heavily on data-driven methods, while neglecting the structural aspects of the model's reasoning capacity. To encourage a more structural generation of CoT steps, we propose a hierarchical generation scheme: we let the LM generate a planning token at the start of each reasoning step, intuitively serving as a high-level plan of the current step, and add thei"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05707","kind":"arxiv","version":4},"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/2310.05707/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":"2310.05707","created_at":"2026-07-05T08:52:51.212369+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.05707v4","created_at":"2026-07-05T08:52:51.212369+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05707","created_at":"2026-07-05T08:52:51.212369+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZKO4P5EBGEBZ","created_at":"2026-07-05T08:52:51.212369+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZKO4P5EBGEBZUX4V","created_at":"2026-07-05T08:52:51.212369+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZKO4P5EB","created_at":"2026-07-05T08:52:51.212369+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00341","citing_title":"DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28600","citing_title":"Transformers Provably Learn to Internalize Chain-of-Thought","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2506.11274","citing_title":"Learning a Continue-Thinking Token for Enhanced Test-Time Scaling","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19773","citing_title":"PR-CAD: Progressive Refinement for Unified Controllable and Faithful Text-to-CAD Generation with Large Language Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08221","citing_title":"NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06769","citing_title":"Training Large Language Models to Reason in a Continuous Latent Space","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08299","citing_title":"SeLaR: Selective Latent Reasoning in Large Language Models","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":50,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT","json":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT.json","graph_json":"https://pith.science/api/pith-number/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/graph.json","events_json":"https://pith.science/api/pith-number/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/events.json","paper":"https://pith.science/paper/ZKO4P5EB"},"agent_actions":{"view_html":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT","download_json":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT.json","view_paper":"https://pith.science/paper/ZKO4P5EB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.05707&json=true","fetch_graph":"https://pith.science/api/pith-number/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/graph.json","fetch_events":"https://pith.science/api/pith-number/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/action/storage_attestation","attest_author":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/action/author_attestation","sign_citation":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/action/citation_signature","submit_replication":"https://pith.science/pith/ZKO4P5EBGEBZUX4VWNYQVZ7JNT/action/replication_record"}},"created_at":"2026-07-05T08:52:51.212369+00:00","updated_at":"2026-07-05T08:52:51.212369+00:00"}