{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HAMJRHDVBSGO4X3REB23KZDYUD","short_pith_number":"pith:HAMJRHDV","schema_version":"1.0","canonical_sha256":"3818989c750c8cee5f712075b56478a0c10342885ff5955d7f8394e64362157e","source":{"kind":"arxiv","id":"2408.12337","version":1},"attestation_state":"computed","paper":{"title":"Fine-tuning Smaller Language Models for Question Answering over Financial Documents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SY","eess.SY"],"primary_cat":"cs.CL","authors_text":"Chetan Harsha, Karmvir Singh Phogat, Sai Akhil Puranam, Shashishekar Ramakrishna, Sridhar Dasaratha","submitted_at":"2024-08-22T12:23:29Z","abstract_excerpt":"Recent research has shown that smaller language models can acquire substantial reasoning abilities when fine-tuned with reasoning exemplars crafted by a significantly larger teacher model. We explore this paradigm for the financial domain, focusing on the challenge of answering questions that require multi-hop numerical reasoning over financial texts. We assess the performance of several smaller models that have been fine-tuned to generate programs that encode the required financial reasoning and calculations. Our findings demonstrate that these fine-tuned smaller models approach the performan"},"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":"2408.12337","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-22T12:23:29Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SY","eess.SY"],"title_canon_sha256":"23b932be3811802e19cef5eb6c28931946b5ca4da4e37fa1475a6d49be2918b0","abstract_canon_sha256":"679d4966a6c68ba151f804f516e805900c8c3502620894266a8223c0d0935eec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:58:07.692857Z","signature_b64":"Gc8jZFcBFTcfglMrpieZN0TJItOHhhOEXaUP63DbKMMLUDmb+Y5QZCgaFuKBy9qoUjOzqj0D+mW+qYz94PPaAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3818989c750c8cee5f712075b56478a0c10342885ff5955d7f8394e64362157e","last_reissued_at":"2026-07-05T08:58:07.692409Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:58:07.692409Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fine-tuning Smaller Language Models for Question Answering over Financial Documents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SY","eess.SY"],"primary_cat":"cs.CL","authors_text":"Chetan Harsha, Karmvir Singh Phogat, Sai Akhil Puranam, Shashishekar Ramakrishna, Sridhar Dasaratha","submitted_at":"2024-08-22T12:23:29Z","abstract_excerpt":"Recent research has shown that smaller language models can acquire substantial reasoning abilities when fine-tuned with reasoning exemplars crafted by a significantly larger teacher model. We explore this paradigm for the financial domain, focusing on the challenge of answering questions that require multi-hop numerical reasoning over financial texts. We assess the performance of several smaller models that have been fine-tuned to generate programs that encode the required financial reasoning and calculations. Our findings demonstrate that these fine-tuned smaller models approach the performan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.12337","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/2408.12337/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":"2408.12337","created_at":"2026-07-05T08:58:07.692471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.12337v1","created_at":"2026-07-05T08:58:07.692471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.12337","created_at":"2026-07-05T08:58:07.692471+00:00"},{"alias_kind":"pith_short_12","alias_value":"HAMJRHDVBSGO","created_at":"2026-07-05T08:58:07.692471+00:00"},{"alias_kind":"pith_short_16","alias_value":"HAMJRHDVBSGO4X3R","created_at":"2026-07-05T08:58:07.692471+00:00"},{"alias_kind":"pith_short_8","alias_value":"HAMJRHDV","created_at":"2026-07-05T08:58:07.692471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.01314","citing_title":"Can Smaller LLMs do better? Unlocking Cross-Domain Potential through Parameter-Efficient Fine-Tuning for Text Summarization","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD","json":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD.json","graph_json":"https://pith.science/api/pith-number/HAMJRHDVBSGO4X3REB23KZDYUD/graph.json","events_json":"https://pith.science/api/pith-number/HAMJRHDVBSGO4X3REB23KZDYUD/events.json","paper":"https://pith.science/paper/HAMJRHDV"},"agent_actions":{"view_html":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD","download_json":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD.json","view_paper":"https://pith.science/paper/HAMJRHDV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.12337&json=true","fetch_graph":"https://pith.science/api/pith-number/HAMJRHDVBSGO4X3REB23KZDYUD/graph.json","fetch_events":"https://pith.science/api/pith-number/HAMJRHDVBSGO4X3REB23KZDYUD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD/action/storage_attestation","attest_author":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD/action/author_attestation","sign_citation":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD/action/citation_signature","submit_replication":"https://pith.science/pith/HAMJRHDVBSGO4X3REB23KZDYUD/action/replication_record"}},"created_at":"2026-07-05T08:58:07.692471+00:00","updated_at":"2026-07-05T08:58:07.692471+00:00"}