{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7276OSWPA4WAZXYSR7X64VF363","short_pith_number":"pith:7276OSWP","schema_version":"1.0","canonical_sha256":"febfe74acf072c0cdf128fefee54bbf6d2d4aa9846c5bfe2589790013bcd6b40","source":{"kind":"arxiv","id":"2602.03491","version":2},"attestation_state":"computed","paper":{"title":"Decoupling Skeleton and Flesh: Efficient Multimodal Table Reasoning with Disentangled Alignment and Structure-aware Guidance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Kehai Chen, Min Zhang, Xiaoqiang Zhou, Xuefeng Bai, Yang Xiang, Yingjie Zhu, Youcheng Pan","submitted_at":"2026-02-03T13:08:31Z","abstract_excerpt":"Reasoning over table images remains challenging for Large Vision-Language Models (LVLMs) due to complex layouts and tightly coupled structure-content information. Existing solutions often depend on expensive supervised training, reinforcement learning, or external tools, limiting efficiency and scalability. This work addresses a key question: how to adapt LVLMs to table reasoning with minimal annotation and no external tools? Specifically, we first introduce DiSCo, a Disentangled Structure-Content alignment framework that explicitly separates structural abstraction from semantic grounding duri"},"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":"2602.03491","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-02-03T13:08:31Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9222d22df30b6a5f396e165acfaef440c2ba8287563affdec47f721f2b3481ea","abstract_canon_sha256":"e28bb53c8d44d08ebbe206ccbfc91641e65c9324cd264c2e5937d79b01393f67"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T02:04:45.907894Z","signature_b64":"VHMBrcoQP7vjYOcT5i7PtbgqIRxILq+ni1XG0kdmEL74Or7jBHm9eG5bBUoYkN6asgG/UkWJGisBXK5Yzd6uCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"febfe74acf072c0cdf128fefee54bbf6d2d4aa9846c5bfe2589790013bcd6b40","last_reissued_at":"2026-05-28T02:04:45.907385Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T02:04:45.907385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Decoupling Skeleton and Flesh: Efficient Multimodal Table Reasoning with Disentangled Alignment and Structure-aware Guidance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Kehai Chen, Min Zhang, Xiaoqiang Zhou, Xuefeng Bai, Yang Xiang, Yingjie Zhu, Youcheng Pan","submitted_at":"2026-02-03T13:08:31Z","abstract_excerpt":"Reasoning over table images remains challenging for Large Vision-Language Models (LVLMs) due to complex layouts and tightly coupled structure-content information. Existing solutions often depend on expensive supervised training, reinforcement learning, or external tools, limiting efficiency and scalability. This work addresses a key question: how to adapt LVLMs to table reasoning with minimal annotation and no external tools? Specifically, we first introduce DiSCo, a Disentangled Structure-Content alignment framework that explicitly separates structural abstraction from semantic grounding duri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.03491","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/2602.03491/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":"2602.03491","created_at":"2026-05-28T02:04:45.907444+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.03491v2","created_at":"2026-05-28T02:04:45.907444+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.03491","created_at":"2026-05-28T02:04:45.907444+00:00"},{"alias_kind":"pith_short_12","alias_value":"7276OSWPA4WA","created_at":"2026-05-28T02:04:45.907444+00:00"},{"alias_kind":"pith_short_16","alias_value":"7276OSWPA4WAZXYS","created_at":"2026-05-28T02:04:45.907444+00:00"},{"alias_kind":"pith_short_8","alias_value":"7276OSWP","created_at":"2026-05-28T02:04:45.907444+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.17982","citing_title":"Mitigating Multimodal Hallucination via Phase-wise Self-reward","ref_index":65,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363","json":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363.json","graph_json":"https://pith.science/api/pith-number/7276OSWPA4WAZXYSR7X64VF363/graph.json","events_json":"https://pith.science/api/pith-number/7276OSWPA4WAZXYSR7X64VF363/events.json","paper":"https://pith.science/paper/7276OSWP"},"agent_actions":{"view_html":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363","download_json":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363.json","view_paper":"https://pith.science/paper/7276OSWP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.03491&json=true","fetch_graph":"https://pith.science/api/pith-number/7276OSWPA4WAZXYSR7X64VF363/graph.json","fetch_events":"https://pith.science/api/pith-number/7276OSWPA4WAZXYSR7X64VF363/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363/action/storage_attestation","attest_author":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363/action/author_attestation","sign_citation":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363/action/citation_signature","submit_replication":"https://pith.science/pith/7276OSWPA4WAZXYSR7X64VF363/action/replication_record"}},"created_at":"2026-05-28T02:04:45.907444+00:00","updated_at":"2026-05-28T02:04:45.907444+00:00"}