{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WDXXXGLW446BC2XZDRSFHW6EUS","short_pith_number":"pith:WDXXXGLW","schema_version":"1.0","canonical_sha256":"b0ef7b9976e73c116af91c6453dbc4a4a87e5181b4ed7dec92c232459c375aa2","source":{"kind":"arxiv","id":"2407.19669","version":2},"attestation_state":"computed","paper":{"title":"mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Baosong Yang, Dingkun Long, Fei Huang, Huan Lin, Jialong Tang, Meishan Zhang, Min Zhang, Pengjun Xie, Wenjie Li, Wen Xie, Xin Zhang, Yanzhao Zhang, Ziqi Dai","submitted_at":"2024-07-29T03:12:28Z","abstract_excerpt":"We present systematic efforts in building long-context multilingual text representation model (TRM) and reranker from scratch for text retrieval. We first introduce a text encoder (base size) enhanced with RoPE and unpadding, pre-trained in a native 8192-token context (longer than 512 of previous multilingual encoders). Then we construct a hybrid TRM and a cross-encoder reranker by contrastive learning. Evaluations show that our text encoder outperforms the same-sized previous state-of-the-art XLM-R. Meanwhile, our TRM and reranker match the performance of large-sized state-of-the-art BGE-M3 m"},"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":"2407.19669","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-07-29T03:12:28Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"4147663170abb3c27758a09e7088921a09fbac34bc18eb491a29c780a52bb08f","abstract_canon_sha256":"f5bb40bb6bcfb868836a5af7ade5537e8a6c99301976ce2e7dda353ee06e2d55"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:20:01.530251Z","signature_b64":"G2L4ZkStffVL/HAcO9yJofYFBZadq6I/SevAxofYiq2povmyNMoDU492rKsAcQCvD2gDkkFWg01lbWwh9zbaCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0ef7b9976e73c116af91c6453dbc4a4a87e5181b4ed7dec92c232459c375aa2","last_reissued_at":"2026-07-05T09:20:01.529818Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:20:01.529818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Baosong Yang, Dingkun Long, Fei Huang, Huan Lin, Jialong Tang, Meishan Zhang, Min Zhang, Pengjun Xie, Wenjie Li, Wen Xie, Xin Zhang, Yanzhao Zhang, Ziqi Dai","submitted_at":"2024-07-29T03:12:28Z","abstract_excerpt":"We present systematic efforts in building long-context multilingual text representation model (TRM) and reranker from scratch for text retrieval. We first introduce a text encoder (base size) enhanced with RoPE and unpadding, pre-trained in a native 8192-token context (longer than 512 of previous multilingual encoders). Then we construct a hybrid TRM and a cross-encoder reranker by contrastive learning. Evaluations show that our text encoder outperforms the same-sized previous state-of-the-art XLM-R. Meanwhile, our TRM and reranker match the performance of large-sized state-of-the-art BGE-M3 m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.19669","kind":"arxiv","version":2},"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/2407.19669/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":"2407.19669","created_at":"2026-07-05T09:20:01.529876+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.19669v2","created_at":"2026-07-05T09:20:01.529876+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.19669","created_at":"2026-07-05T09:20:01.529876+00:00"},{"alias_kind":"pith_short_12","alias_value":"WDXXXGLW446B","created_at":"2026-07-05T09:20:01.529876+00:00"},{"alias_kind":"pith_short_16","alias_value":"WDXXXGLW446BC2XZ","created_at":"2026-07-05T09:20:01.529876+00:00"},{"alias_kind":"pith_short_8","alias_value":"WDXXXGLW","created_at":"2026-07-05T09:20:01.529876+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22778","citing_title":"HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions","ref_index":148,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29992","citing_title":"Adapting Multilingual Embedding Models to Turkish via Cross-Lingual Tokenizer Surgery and Offline Distillation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2507.08480","citing_title":"Improving Korean-English Cross-Lingual Retrieval: A Data-Centric Study of Language Composition and Model Merging","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13415","citing_title":"KIT-TIP-NLP at MultiPride: Continual Learning with Multilingual Foundation Model","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2507.07847","citing_title":"From Ambiguity to Accuracy: The Transformative Effect of Coreference Resolution on Retrieval-Augmented Generation systems","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2510.01409","citing_title":"OntoLogX: Ontology-Guided Knowledge Graph Extraction from Cybersecurity Logs with Large Language Models","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13415","citing_title":"KIT-TIP-NLP at MultiPride: Continual Learning with Multilingual Foundation Model","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13521","citing_title":"Granite Embedding Multilingual R2 Models","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2502.02737","citing_title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","ref_index":252,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05767","citing_title":"Priming, Path-dependence, and Plasticity: Understanding the molding of user-LLM interaction and its implications from (many) chat logs in the wild","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20666","citing_title":"ORPHEAS: A Cross-Lingual Greek-English Embedding Model for Retrieval-Augmented Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12081","citing_title":"Human-Inspired Context-Selective Multimodal Memory for Social Robots","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17257","citing_title":"REZE: Representation Regularization for Domain-adaptive Text Embedding Pre-finetuning","ref_index":68,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS","json":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS.json","graph_json":"https://pith.science/api/pith-number/WDXXXGLW446BC2XZDRSFHW6EUS/graph.json","events_json":"https://pith.science/api/pith-number/WDXXXGLW446BC2XZDRSFHW6EUS/events.json","paper":"https://pith.science/paper/WDXXXGLW"},"agent_actions":{"view_html":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS","download_json":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS.json","view_paper":"https://pith.science/paper/WDXXXGLW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.19669&json=true","fetch_graph":"https://pith.science/api/pith-number/WDXXXGLW446BC2XZDRSFHW6EUS/graph.json","fetch_events":"https://pith.science/api/pith-number/WDXXXGLW446BC2XZDRSFHW6EUS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS/action/storage_attestation","attest_author":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS/action/author_attestation","sign_citation":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS/action/citation_signature","submit_replication":"https://pith.science/pith/WDXXXGLW446BC2XZDRSFHW6EUS/action/replication_record"}},"created_at":"2026-07-05T09:20:01.529876+00:00","updated_at":"2026-07-05T09:20:01.529876+00:00"}