{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SXTFJDHDD7CEGS47DIB6VKJ4MJ","short_pith_number":"pith:SXTFJDHD","schema_version":"1.0","canonical_sha256":"95e6548ce31fc4434b9f1a03eaa93c624bf38fb77730680e048c0b93b5925f84","source":{"kind":"arxiv","id":"2402.10986","version":3},"attestation_state":"computed","paper":{"title":"FinTral: A Family of GPT-4 Level Multimodal Financial Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"El Moatez Billah Nagoudi, Gagan Bhatia, Hasan Cavusoglu, Muhammad Abdul-Mageed","submitted_at":"2024-02-16T05:05:12Z","abstract_excerpt":"We introduce FinTral, a suite of state-of-the-art multimodal large language models (LLMs) built upon the Mistral-7b model and tailored for financial analysis. FinTral integrates textual, numerical, tabular, and image data. We enhance FinTral with domain-specific pretraining, instruction fine-tuning, and RLAIF training by exploiting a large collection of textual and visual datasets we curate for this work. We also introduce an extensive benchmark featuring nine tasks and 25 datasets for evaluation, including hallucinations in the financial domain. Our FinTral model trained with direct preferenc"},"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":"2402.10986","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T05:05:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5bb9375ae09b3a65d6c69e7e5e9c3a3667d24826753a99d7537fef4cd6e199df","abstract_canon_sha256":"325bfcf11bcb1d20fbf2dc4a703b07105c3f238f2d451d143bcb476f22020217"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:31:48.432134Z","signature_b64":"am1tnIMKGw4+FCatJGVj39aG6JdAI7fH42KV0ThO7eBGsTP4HdM+ghxxMN8isuJj9hwk6pT4dAMXAUyb/c2LCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95e6548ce31fc4434b9f1a03eaa93c624bf38fb77730680e048c0b93b5925f84","last_reissued_at":"2026-07-05T08:31:48.431647Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:31:48.431647Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FinTral: A Family of GPT-4 Level Multimodal Financial Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"El Moatez Billah Nagoudi, Gagan Bhatia, Hasan Cavusoglu, Muhammad Abdul-Mageed","submitted_at":"2024-02-16T05:05:12Z","abstract_excerpt":"We introduce FinTral, a suite of state-of-the-art multimodal large language models (LLMs) built upon the Mistral-7b model and tailored for financial analysis. FinTral integrates textual, numerical, tabular, and image data. We enhance FinTral with domain-specific pretraining, instruction fine-tuning, and RLAIF training by exploiting a large collection of textual and visual datasets we curate for this work. We also introduce an extensive benchmark featuring nine tasks and 25 datasets for evaluation, including hallucinations in the financial domain. Our FinTral model trained with direct preferenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10986","kind":"arxiv","version":3},"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/2402.10986/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":"2402.10986","created_at":"2026-07-05T08:31:48.431698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.10986v3","created_at":"2026-07-05T08:31:48.431698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10986","created_at":"2026-07-05T08:31:48.431698+00:00"},{"alias_kind":"pith_short_12","alias_value":"SXTFJDHDD7CE","created_at":"2026-07-05T08:31:48.431698+00:00"},{"alias_kind":"pith_short_16","alias_value":"SXTFJDHDD7CEGS47","created_at":"2026-07-05T08:31:48.431698+00:00"},{"alias_kind":"pith_short_8","alias_value":"SXTFJDHD","created_at":"2026-07-05T08:31:48.431698+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2406.12009","citing_title":"FinTruthQA: A Benchmark for AI-Driven Financial Disclosure Quality Assessment in Investor -- Firm Interactions","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2503.22693","citing_title":"Bridging Language Models and Financial Analysis","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2506.11512","citing_title":"From Time Series Analysis to Question Answering: A Survey in the LLM Era","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ","json":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ.json","graph_json":"https://pith.science/api/pith-number/SXTFJDHDD7CEGS47DIB6VKJ4MJ/graph.json","events_json":"https://pith.science/api/pith-number/SXTFJDHDD7CEGS47DIB6VKJ4MJ/events.json","paper":"https://pith.science/paper/SXTFJDHD"},"agent_actions":{"view_html":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ","download_json":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ.json","view_paper":"https://pith.science/paper/SXTFJDHD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.10986&json=true","fetch_graph":"https://pith.science/api/pith-number/SXTFJDHDD7CEGS47DIB6VKJ4MJ/graph.json","fetch_events":"https://pith.science/api/pith-number/SXTFJDHDD7CEGS47DIB6VKJ4MJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ/action/storage_attestation","attest_author":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ/action/author_attestation","sign_citation":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ/action/citation_signature","submit_replication":"https://pith.science/pith/SXTFJDHDD7CEGS47DIB6VKJ4MJ/action/replication_record"}},"created_at":"2026-07-05T08:31:48.431698+00:00","updated_at":"2026-07-05T08:31:48.431698+00:00"}