{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HISFQYN63XKLXFU3YGOOURNY6M","short_pith_number":"pith:HISFQYN6","schema_version":"1.0","canonical_sha256":"3a245861beddd4bb969bc19cea45b8f314444c30decc05a0dd9d563a82e85e43","source":{"kind":"arxiv","id":"2405.18414","version":1},"attestation_state":"computed","paper":{"title":"Don't Forget to Connect! Improving RAG with Graph-based Reranking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SI"],"primary_cat":"cs.CL","authors_text":"Anton Tsitsulin, Bahare Fatemi, Bryan Perozzi, Jialin Dong, Lin F. Yang","submitted_at":"2024-05-28T17:56:46Z","abstract_excerpt":"Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question context. But what about when a document has partial information, or less obvious connections to the context? And how should we reason about connections between documents? In this work, we seek to answer these two core questions about RAG generation. We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reade"},"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":"2405.18414","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-28T17:56:46Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SI"],"title_canon_sha256":"7474fc5a338c7e072d7f6c4d9ff73eaf9c14aacf0b6d15d33ce34507e4dcf30d","abstract_canon_sha256":"afbd2841f16c2169de2a936e7ac98cd5da0dd5ed6278fd585dbe217bdd6830b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:19.889216Z","signature_b64":"I/Q0fyp1s8ODec2pH95JdHKReoN3dzmz8EzaMeWKggh5qz0yfrm+PwereiwAERMmjSCmmJMWSRMKgsTYqzeUBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a245861beddd4bb969bc19cea45b8f314444c30decc05a0dd9d563a82e85e43","last_reissued_at":"2026-07-05T08:24:19.888734Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:19.888734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Don't Forget to Connect! Improving RAG with Graph-based Reranking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SI"],"primary_cat":"cs.CL","authors_text":"Anton Tsitsulin, Bahare Fatemi, Bryan Perozzi, Jialin Dong, Lin F. Yang","submitted_at":"2024-05-28T17:56:46Z","abstract_excerpt":"Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question context. But what about when a document has partial information, or less obvious connections to the context? And how should we reason about connections between documents? In this work, we seek to answer these two core questions about RAG generation. We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reade"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.18414","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/2405.18414/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":"2405.18414","created_at":"2026-07-05T08:24:19.888806+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.18414v1","created_at":"2026-07-05T08:24:19.888806+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.18414","created_at":"2026-07-05T08:24:19.888806+00:00"},{"alias_kind":"pith_short_12","alias_value":"HISFQYN63XKL","created_at":"2026-07-05T08:24:19.888806+00:00"},{"alias_kind":"pith_short_16","alias_value":"HISFQYN63XKLXFU3","created_at":"2026-07-05T08:24:19.888806+00:00"},{"alias_kind":"pith_short_8","alias_value":"HISFQYN6","created_at":"2026-07-05T08:24:19.888806+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14236","citing_title":"Active Learners as Efficient PRP Rerankers","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2501.00309","citing_title":"Retrieval-Augmented Generation with Graphs (GraphRAG)","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14192","citing_title":"Why Retrieval-Augmented Generation Fails: A Graph Perspective","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14236","citing_title":"Active Learners as Efficient PRP Rerankers","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27852","citing_title":"NeocorRAG: Less Irrelevant Information, More Explicit Evidence, and More Effective Recall via Evidence Chains","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07517","citing_title":"LARAG: Link-Aware Retrieval Strategy for RAG Systems in Hyperlinked Technical Documentation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17458","citing_title":"EHRAG: Bridging Semantic Gaps in Lightweight GraphRAG via Hybrid Hypergraph Construction and Retrieval","ref_index":152,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M","json":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M.json","graph_json":"https://pith.science/api/pith-number/HISFQYN63XKLXFU3YGOOURNY6M/graph.json","events_json":"https://pith.science/api/pith-number/HISFQYN63XKLXFU3YGOOURNY6M/events.json","paper":"https://pith.science/paper/HISFQYN6"},"agent_actions":{"view_html":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M","download_json":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M.json","view_paper":"https://pith.science/paper/HISFQYN6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.18414&json=true","fetch_graph":"https://pith.science/api/pith-number/HISFQYN63XKLXFU3YGOOURNY6M/graph.json","fetch_events":"https://pith.science/api/pith-number/HISFQYN63XKLXFU3YGOOURNY6M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M/action/storage_attestation","attest_author":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M/action/author_attestation","sign_citation":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M/action/citation_signature","submit_replication":"https://pith.science/pith/HISFQYN63XKLXFU3YGOOURNY6M/action/replication_record"}},"created_at":"2026-07-05T08:24:19.888806+00:00","updated_at":"2026-07-05T08:24:19.888806+00:00"}