{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2ZJFTYBJXRLIKKTLOGSRDEWIJL","short_pith_number":"pith:2ZJFTYBJ","schema_version":"1.0","canonical_sha256":"d65259e029bc56852a6b71a51192c84ad4fa040cb676406192de034ef7d9d423","source":{"kind":"arxiv","id":"2606.28916","version":1},"attestation_state":"computed","paper":{"title":"Latent Bridges for Multi-Table Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DB"],"primary_cat":"cs.CL","authors_text":"Floris Geerts, Paolo Papotti, Simone Varriale, Tamara Cucumides","submitted_at":"2026-06-27T13:48:20Z","abstract_excerpt":"We introduce GRAB, a constructor-encoder-bridge pipeline for table question answering. Our method lifts relational data into an heterogeneous graph, encodes it via message passing, and transfers the signals to an LLM through a small set of query-conditioned latent tokens. This provides the LLM with a compact, task-relevant structural representation together with the flattened text. Crucially, the LLM remains strictly frozen to preserve its general reasoning capabilities; we train only the lightweight graph encoder and latent bridge (91M parameters), allowing the entire pipeline to be trained e"},"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":"2606.28916","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-27T13:48:20Z","cross_cats_sorted":["cs.AI","cs.DB"],"title_canon_sha256":"a00af6b6dc4aa1915771d2db0d88f3e99167fa5895a1b4354140c829d010df66","abstract_canon_sha256":"f5394d72ac76fce356d1a6b08d921b4c6c7dac19d3aa15bc0f060c31db4271d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:17:45.125190Z","signature_b64":"MN5RC5E1FBATQyJHK5ReN/6RtwSexg3o3wnqMfIh/Ao6KwuwWbXJGi6zpLrWB8994yA861UWhm+OJBZsiUgFDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d65259e029bc56852a6b71a51192c84ad4fa040cb676406192de034ef7d9d423","last_reissued_at":"2026-06-30T01:17:45.124388Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:17:45.124388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Latent Bridges for Multi-Table Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DB"],"primary_cat":"cs.CL","authors_text":"Floris Geerts, Paolo Papotti, Simone Varriale, Tamara Cucumides","submitted_at":"2026-06-27T13:48:20Z","abstract_excerpt":"We introduce GRAB, a constructor-encoder-bridge pipeline for table question answering. Our method lifts relational data into an heterogeneous graph, encodes it via message passing, and transfers the signals to an LLM through a small set of query-conditioned latent tokens. This provides the LLM with a compact, task-relevant structural representation together with the flattened text. Crucially, the LLM remains strictly frozen to preserve its general reasoning capabilities; we train only the lightweight graph encoder and latent bridge (91M parameters), allowing the entire pipeline to be trained e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28916","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/2606.28916/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":"2606.28916","created_at":"2026-06-30T01:17:45.124537+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.28916v1","created_at":"2026-06-30T01:17:45.124537+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28916","created_at":"2026-06-30T01:17:45.124537+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ZJFTYBJXRLI","created_at":"2026-06-30T01:17:45.124537+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ZJFTYBJXRLIKKTL","created_at":"2026-06-30T01:17:45.124537+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ZJFTYBJ","created_at":"2026-06-30T01:17:45.124537+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/2ZJFTYBJXRLIKKTLOGSRDEWIJL","json":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL.json","graph_json":"https://pith.science/api/pith-number/2ZJFTYBJXRLIKKTLOGSRDEWIJL/graph.json","events_json":"https://pith.science/api/pith-number/2ZJFTYBJXRLIKKTLOGSRDEWIJL/events.json","paper":"https://pith.science/paper/2ZJFTYBJ"},"agent_actions":{"view_html":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL","download_json":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL.json","view_paper":"https://pith.science/paper/2ZJFTYBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.28916&json=true","fetch_graph":"https://pith.science/api/pith-number/2ZJFTYBJXRLIKKTLOGSRDEWIJL/graph.json","fetch_events":"https://pith.science/api/pith-number/2ZJFTYBJXRLIKKTLOGSRDEWIJL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL/action/storage_attestation","attest_author":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL/action/author_attestation","sign_citation":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL/action/citation_signature","submit_replication":"https://pith.science/pith/2ZJFTYBJXRLIKKTLOGSRDEWIJL/action/replication_record"}},"created_at":"2026-06-30T01:17:45.124537+00:00","updated_at":"2026-06-30T01:17:45.124537+00:00"}