{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SA4BCT6TRO6V7NNPUXIRT3VKJM","short_pith_number":"pith:SA4BCT6T","schema_version":"1.0","canonical_sha256":"9038114fd38bbd5fb5afa5d119eeaa4b1649d90d31859258ed90b34c0a653aba","source":{"kind":"arxiv","id":"2605.30587","version":1},"attestation_state":"computed","paper":{"title":"ReGuLaR: Relation-Grounded Latent Reasoning for Large Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Karthik Somayaji N.S, Peng Li, Zihu Wang","submitted_at":"2026-05-28T21:34:40Z","abstract_excerpt":"Chain-of-thought (CoT) reasoning has significantly improved the reasoning ability of large vision-language models (LVLMs) by verbalizing intermediate reasoning steps in natural language. However, such discrete textual rationales are often insufficient for encoding continuous visual evidence. Recent work addresses this limitation by moving reasoning into continuous latent space. Despite promising progress, existing methods leave latent reasoning insufficiently connected to the compositional and relational structure of visual evidence. To address this gap, we introduce ReGuLaR, a relation ground"},"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":"2605.30587","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-28T21:34:40Z","cross_cats_sorted":[],"title_canon_sha256":"e549d8bce616c213b4aa50f95d9fd4dc2a339fe8706c74a8ddb3c62184131e3c","abstract_canon_sha256":"fe7535f4c7ec4ad03906967eca6de66e0e0bb1660ec4482bb4798bfefdeb3596"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:02.664168Z","signature_b64":"Ib64ycgJQI04EI+WWVKKRV37tatw7prNgojOuzYLu98VzNfVP9Q2UjACsXIsDukmRTEDAcNAaAIFfF/UFVUpCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9038114fd38bbd5fb5afa5d119eeaa4b1649d90d31859258ed90b34c0a653aba","last_reissued_at":"2026-06-01T01:03:02.663289Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:02.663289Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReGuLaR: Relation-Grounded Latent Reasoning for Large Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Karthik Somayaji N.S, Peng Li, Zihu Wang","submitted_at":"2026-05-28T21:34:40Z","abstract_excerpt":"Chain-of-thought (CoT) reasoning has significantly improved the reasoning ability of large vision-language models (LVLMs) by verbalizing intermediate reasoning steps in natural language. However, such discrete textual rationales are often insufficient for encoding continuous visual evidence. Recent work addresses this limitation by moving reasoning into continuous latent space. Despite promising progress, existing methods leave latent reasoning insufficiently connected to the compositional and relational structure of visual evidence. To address this gap, we introduce ReGuLaR, a relation ground"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30587","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/2605.30587/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":"2605.30587","created_at":"2026-06-01T01:03:02.663456+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30587v1","created_at":"2026-06-01T01:03:02.663456+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30587","created_at":"2026-06-01T01:03:02.663456+00:00"},{"alias_kind":"pith_short_12","alias_value":"SA4BCT6TRO6V","created_at":"2026-06-01T01:03:02.663456+00:00"},{"alias_kind":"pith_short_16","alias_value":"SA4BCT6TRO6V7NNP","created_at":"2026-06-01T01:03:02.663456+00:00"},{"alias_kind":"pith_short_8","alias_value":"SA4BCT6T","created_at":"2026-06-01T01:03:02.663456+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/SA4BCT6TRO6V7NNPUXIRT3VKJM","json":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM.json","graph_json":"https://pith.science/api/pith-number/SA4BCT6TRO6V7NNPUXIRT3VKJM/graph.json","events_json":"https://pith.science/api/pith-number/SA4BCT6TRO6V7NNPUXIRT3VKJM/events.json","paper":"https://pith.science/paper/SA4BCT6T"},"agent_actions":{"view_html":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM","download_json":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM.json","view_paper":"https://pith.science/paper/SA4BCT6T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30587&json=true","fetch_graph":"https://pith.science/api/pith-number/SA4BCT6TRO6V7NNPUXIRT3VKJM/graph.json","fetch_events":"https://pith.science/api/pith-number/SA4BCT6TRO6V7NNPUXIRT3VKJM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM/action/storage_attestation","attest_author":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM/action/author_attestation","sign_citation":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM/action/citation_signature","submit_replication":"https://pith.science/pith/SA4BCT6TRO6V7NNPUXIRT3VKJM/action/replication_record"}},"created_at":"2026-06-01T01:03:02.663456+00:00","updated_at":"2026-06-01T01:03:02.663456+00:00"}