{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AU6W3OU2JO4TCC6NDUASZUFUWM","short_pith_number":"pith:AU6W3OU2","schema_version":"1.0","canonical_sha256":"053d6dba9a4bb9310bcd1d012cd0b4b3235e559351aa7badb953dbdb0f3315e4","source":{"kind":"arxiv","id":"2606.08605","version":1},"attestation_state":"computed","paper":{"title":"Multilingual Fact-Checking at Scale: Fine-Tuned Compact Models vs LLMs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pratuat Amatya, Vinay Setty","submitted_at":"2026-06-07T12:33:43Z","abstract_excerpt":"We present a multilingual fact-checking system deployed at Factiverse, designed for high-throughput and low-latency operation across diverse languages. The system follows a modular pipeline with three stages: claim detection, evidence retrieval and re-ranking, and veracity prediction. We fine-tune XLM-RoBERTa-Large for claim detection, mmBERT-base for three-label stance classification (Supports/Refutes/Mixed), and a SetFit-based multilingual re-ranker for claim--evidence matching. We compare these components against strong LLM baselines, including GPT-5.2, Claude Opus~4.6, and Qwen3-8b. Experi"},"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":"2606.08605","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-07T12:33:43Z","cross_cats_sorted":[],"title_canon_sha256":"276e64b69db04199f681fdd4868db0e1fb713b36e8d666628439e065755136fb","abstract_canon_sha256":"27c4e9915fd5377e96a73ca4be2eb53dbbb932d87ce10cfdf86a4305d5ff3ee4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:41.261846Z","signature_b64":"fqiLhtOZaL4crEDK92D/2SBKsUSB+D6E47lZoNqlToJPlrsO9Ui8fNU4WZrut8pmQoLsyuyPa4ESA4iGpzj+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"053d6dba9a4bb9310bcd1d012cd0b4b3235e559351aa7badb953dbdb0f3315e4","last_reissued_at":"2026-06-09T01:05:41.261324Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:41.261324Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Fact-Checking at Scale: Fine-Tuned Compact Models vs LLMs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pratuat Amatya, Vinay Setty","submitted_at":"2026-06-07T12:33:43Z","abstract_excerpt":"We present a multilingual fact-checking system deployed at Factiverse, designed for high-throughput and low-latency operation across diverse languages. The system follows a modular pipeline with three stages: claim detection, evidence retrieval and re-ranking, and veracity prediction. We fine-tune XLM-RoBERTa-Large for claim detection, mmBERT-base for three-label stance classification (Supports/Refutes/Mixed), and a SetFit-based multilingual re-ranker for claim--evidence matching. We compare these components against strong LLM baselines, including GPT-5.2, Claude Opus~4.6, and Qwen3-8b. Experi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08605","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/2606.08605/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":"2606.08605","created_at":"2026-06-09T01:05:41.261407+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08605v1","created_at":"2026-06-09T01:05:41.261407+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08605","created_at":"2026-06-09T01:05:41.261407+00:00"},{"alias_kind":"pith_short_12","alias_value":"AU6W3OU2JO4T","created_at":"2026-06-09T01:05:41.261407+00:00"},{"alias_kind":"pith_short_16","alias_value":"AU6W3OU2JO4TCC6N","created_at":"2026-06-09T01:05:41.261407+00:00"},{"alias_kind":"pith_short_8","alias_value":"AU6W3OU2","created_at":"2026-06-09T01:05:41.261407+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/AU6W3OU2JO4TCC6NDUASZUFUWM","json":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM.json","graph_json":"https://pith.science/api/pith-number/AU6W3OU2JO4TCC6NDUASZUFUWM/graph.json","events_json":"https://pith.science/api/pith-number/AU6W3OU2JO4TCC6NDUASZUFUWM/events.json","paper":"https://pith.science/paper/AU6W3OU2"},"agent_actions":{"view_html":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM","download_json":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM.json","view_paper":"https://pith.science/paper/AU6W3OU2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08605&json=true","fetch_graph":"https://pith.science/api/pith-number/AU6W3OU2JO4TCC6NDUASZUFUWM/graph.json","fetch_events":"https://pith.science/api/pith-number/AU6W3OU2JO4TCC6NDUASZUFUWM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM/action/storage_attestation","attest_author":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM/action/author_attestation","sign_citation":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM/action/citation_signature","submit_replication":"https://pith.science/pith/AU6W3OU2JO4TCC6NDUASZUFUWM/action/replication_record"}},"created_at":"2026-06-09T01:05:41.261407+00:00","updated_at":"2026-06-09T01:05:41.261407+00:00"}