{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2AQI6GAAMRFD4G3AV2WZVYGOSR","short_pith_number":"pith:2AQI6GAA","schema_version":"1.0","canonical_sha256":"d0208f1800644a3e1b60aead9ae0ce947b8f4ab6069007591422d2ab8a5e46a1","source":{"kind":"arxiv","id":"2607.19691","version":1},"attestation_state":"computed","paper":{"title":"SLPO: Scaling Latent Reasoning via a Surrogate Policy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Runyang You, Wenjie Li, Yongqi Li, Zhiyuan Liu","submitted_at":"2026-07-22T02:45:08Z","abstract_excerpt":"Reinforcement learning with verifiable rewards has become the predominant recipe for eliciting test-time scaling in explicit Chain-of-Thought reasoners. Yet this scaling path remains computationally costly, since every intermediate step must be decoded as a language token. Latent reasoning instead carries intermediate computation as continuous vectors and already matches or surpasses explicit CoT at far shorter horizons. Despite this promise, latent reasoners remain largely imitation-bound, while explicit CoT has already moved past imitation via outcome-reward RL. Latent trajectories lack a tr"},"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":"2607.19691","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-22T02:45:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d03c1337625914fbbe1f98661ab958c937827091c8008f97ed1fda11aa113d4e","abstract_canon_sha256":"783ed84727f7fa69c51532a229374f47a39ce2f14431435fa88dc4278a15dc2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-23T00:24:04.063937Z","signature_b64":"Qk64AC9p9CRD3pV0YRQeZ0vofCwXWbMcx6/SW8uYdtUcnE1i1Ilr3WDWHjp5MkLoArVtfk/4UNWchaXMSdEtAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0208f1800644a3e1b60aead9ae0ce947b8f4ab6069007591422d2ab8a5e46a1","last_reissued_at":"2026-07-23T00:24:04.063043Z","signature_status":"signed_v1","first_computed_at":"2026-07-23T00:24:04.063043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SLPO: Scaling Latent Reasoning via a Surrogate Policy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Runyang You, Wenjie Li, Yongqi Li, Zhiyuan Liu","submitted_at":"2026-07-22T02:45:08Z","abstract_excerpt":"Reinforcement learning with verifiable rewards has become the predominant recipe for eliciting test-time scaling in explicit Chain-of-Thought reasoners. Yet this scaling path remains computationally costly, since every intermediate step must be decoded as a language token. Latent reasoning instead carries intermediate computation as continuous vectors and already matches or surpasses explicit CoT at far shorter horizons. Despite this promise, latent reasoners remain largely imitation-bound, while explicit CoT has already moved past imitation via outcome-reward RL. Latent trajectories lack a tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.19691","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/2607.19691/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":"2607.19691","created_at":"2026-07-23T00:24:04.063495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.19691v1","created_at":"2026-07-23T00:24:04.063495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.19691","created_at":"2026-07-23T00:24:04.063495+00:00"},{"alias_kind":"pith_short_12","alias_value":"2AQI6GAAMRFD","created_at":"2026-07-23T00:24:04.063495+00:00"},{"alias_kind":"pith_short_16","alias_value":"2AQI6GAAMRFD4G3A","created_at":"2026-07-23T00:24:04.063495+00:00"},{"alias_kind":"pith_short_8","alias_value":"2AQI6GAA","created_at":"2026-07-23T00:24:04.063495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR","json":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR.json","graph_json":"https://pith.science/api/pith-number/2AQI6GAAMRFD4G3AV2WZVYGOSR/graph.json","events_json":"https://pith.science/api/pith-number/2AQI6GAAMRFD4G3AV2WZVYGOSR/events.json","paper":"https://pith.science/paper/2AQI6GAA"},"agent_actions":{"view_html":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR","download_json":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR.json","view_paper":"https://pith.science/paper/2AQI6GAA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.19691&json=true","fetch_graph":"https://pith.science/api/pith-number/2AQI6GAAMRFD4G3AV2WZVYGOSR/graph.json","fetch_events":"https://pith.science/api/pith-number/2AQI6GAAMRFD4G3AV2WZVYGOSR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR/action/storage_attestation","attest_author":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR/action/author_attestation","sign_citation":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR/action/citation_signature","submit_replication":"https://pith.science/pith/2AQI6GAAMRFD4G3AV2WZVYGOSR/action/replication_record"}},"created_at":"2026-07-23T00:24:04.063495+00:00","updated_at":"2026-07-23T00:24:04.063495+00:00"}