{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6NWK3NSQ7Y7J7XRZFK5IYB24ZP","short_pith_number":"pith:6NWK3NSQ","schema_version":"1.0","canonical_sha256":"f36cadb650fe3e9fde392aba8c075ccbf60f207ec19170b15f2430b6dee2b59e","source":{"kind":"arxiv","id":"2309.08922","version":1},"attestation_state":"computed","paper":{"title":"Multimodal Multi-Hop Question Answering Through a Conversation Between Tools and Efficiently Finetuned Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bang Liu, Hossein Rajabzadeh, Hyock Ju Kwon, Suyuchen Wang","submitted_at":"2023-09-16T08:22:22Z","abstract_excerpt":"We employ a tool-interacting divide-and-conquer strategy enabling large language models (LLMs) to answer complex multimodal multi-hop questions. In particular, we harness the power of large language models to divide a given multimodal multi-hop question into unimodal single-hop sub-questions to be answered by the appropriate tool from a predefined set of tools. After all corresponding tools provide the LLM with their answers, the LLM generates the next relevant unimodal single-hop question. To increase the reasoning ability of LLMs, we prompt chatGPT to generate a tool-interacting divide-and-c"},"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":"2309.08922","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-16T08:22:22Z","cross_cats_sorted":[],"title_canon_sha256":"1c3e059b861f40e15c6d8bb60819bcda865582871573dc508a1c178f3ebac0ce","abstract_canon_sha256":"4a83cdd6c115ed9bb9ca06d7b3a30ba36ee6c2f5c4b1748c9b20162d0edcacea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:51:31.611189Z","signature_b64":"xNpITv3RlGBJCfxIMKRyWgJQFzzFlp6z83IQuLnCihm+Q7JP/TLOojemVlXoBePmq/9fKi8Ty2R9fRgRtR1EDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f36cadb650fe3e9fde392aba8c075ccbf60f207ec19170b15f2430b6dee2b59e","last_reissued_at":"2026-07-05T06:51:31.610659Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:51:31.610659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Multi-Hop Question Answering Through a Conversation Between Tools and Efficiently Finetuned Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bang Liu, Hossein Rajabzadeh, Hyock Ju Kwon, Suyuchen Wang","submitted_at":"2023-09-16T08:22:22Z","abstract_excerpt":"We employ a tool-interacting divide-and-conquer strategy enabling large language models (LLMs) to answer complex multimodal multi-hop questions. In particular, we harness the power of large language models to divide a given multimodal multi-hop question into unimodal single-hop sub-questions to be answered by the appropriate tool from a predefined set of tools. After all corresponding tools provide the LLM with their answers, the LLM generates the next relevant unimodal single-hop question. To increase the reasoning ability of LLMs, we prompt chatGPT to generate a tool-interacting divide-and-c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.08922","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/2309.08922/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":"2309.08922","created_at":"2026-07-05T06:51:31.610737+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.08922v1","created_at":"2026-07-05T06:51:31.610737+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.08922","created_at":"2026-07-05T06:51:31.610737+00:00"},{"alias_kind":"pith_short_12","alias_value":"6NWK3NSQ7Y7J","created_at":"2026-07-05T06:51:31.610737+00:00"},{"alias_kind":"pith_short_16","alias_value":"6NWK3NSQ7Y7J7XRZ","created_at":"2026-07-05T06:51:31.610737+00:00"},{"alias_kind":"pith_short_8","alias_value":"6NWK3NSQ","created_at":"2026-07-05T06:51:31.610737+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.18027","citing_title":"PDF Retrieval Augmented Question Answering","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP","json":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP.json","graph_json":"https://pith.science/api/pith-number/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/graph.json","events_json":"https://pith.science/api/pith-number/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/events.json","paper":"https://pith.science/paper/6NWK3NSQ"},"agent_actions":{"view_html":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP","download_json":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP.json","view_paper":"https://pith.science/paper/6NWK3NSQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.08922&json=true","fetch_graph":"https://pith.science/api/pith-number/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/graph.json","fetch_events":"https://pith.science/api/pith-number/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/action/storage_attestation","attest_author":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/action/author_attestation","sign_citation":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/action/citation_signature","submit_replication":"https://pith.science/pith/6NWK3NSQ7Y7J7XRZFK5IYB24ZP/action/replication_record"}},"created_at":"2026-07-05T06:51:31.610737+00:00","updated_at":"2026-07-05T06:51:31.610737+00:00"}