{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OFLWBQIJK6CR44E5KACTURL34G","short_pith_number":"pith:OFLWBQIJ","schema_version":"1.0","canonical_sha256":"715760c10957851e709d50053a457be1a84896a106f2a7819127f934eeb44505","source":{"kind":"arxiv","id":"2407.15831","version":2},"attestation_state":"computed","paper":{"title":"NV-Retriever: Improving text embedding models with effective hard-negative mining","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Benedikt Schifferer, Even Oldridge, Gabriel de Souza P. Moreira, Mengyao Xu, Radek Osmulski, Ronay Ak","submitted_at":"2024-07-22T17:50:31Z","abstract_excerpt":"Text embedding models have been popular for information retrieval applications such as semantic search and Question-Answering systems based on Retrieval-Augmented Generation (RAG). Those models are typically Transformer models that are fine-tuned with contrastive learning objectives. One of the challenging aspects of fine-tuning embedding models is the selection of high quality hard-negative passages for contrastive learning. In this paper we introduce a family of positive-aware mining methods that use the positive relevance score as an anchor for effective false negative removal, leading to f"},"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":"2407.15831","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.IR","submitted_at":"2024-07-22T17:50:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4f00e2e655e6ac25e7dc764f17d9c7fec2f5e612b4cc2b7fcabf7fd2b1fcf87b","abstract_canon_sha256":"3933d84e3ebf6fa4b780076677e59e82059375c141bd00e779df461ba7c1769e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:42.085363Z","signature_b64":"C9LrZmZqGPoo7FGRe+WxHhFDUC6eJofUUxTP+yMsjYssFQF5FXhop5W83lx0dUcxMsUnuFsty+v6XOwUlCxDDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"715760c10957851e709d50053a457be1a84896a106f2a7819127f934eeb44505","last_reissued_at":"2026-07-05T10:10:42.084850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:42.084850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NV-Retriever: Improving text embedding models with effective hard-negative mining","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Benedikt Schifferer, Even Oldridge, Gabriel de Souza P. Moreira, Mengyao Xu, Radek Osmulski, Ronay Ak","submitted_at":"2024-07-22T17:50:31Z","abstract_excerpt":"Text embedding models have been popular for information retrieval applications such as semantic search and Question-Answering systems based on Retrieval-Augmented Generation (RAG). Those models are typically Transformer models that are fine-tuned with contrastive learning objectives. One of the challenging aspects of fine-tuning embedding models is the selection of high quality hard-negative passages for contrastive learning. In this paper we introduce a family of positive-aware mining methods that use the positive relevance score as an anchor for effective false negative removal, leading to f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.15831","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/2407.15831/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":"2407.15831","created_at":"2026-07-05T10:10:42.084905+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.15831v2","created_at":"2026-07-05T10:10:42.084905+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.15831","created_at":"2026-07-05T10:10:42.084905+00:00"},{"alias_kind":"pith_short_12","alias_value":"OFLWBQIJK6CR","created_at":"2026-07-05T10:10:42.084905+00:00"},{"alias_kind":"pith_short_16","alias_value":"OFLWBQIJK6CR44E5","created_at":"2026-07-05T10:10:42.084905+00:00"},{"alias_kind":"pith_short_8","alias_value":"OFLWBQIJ","created_at":"2026-07-05T10:10:42.084905+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06036","citing_title":"Learn to Pool: Lightweight Fine-Tuning for Flexible Multi-Vector Compression","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20369","citing_title":"CATCH-ME if you RAG: a dataset of Contextually Annotated multi-Turn Counterspeech against Hate and Misinformation Exchanges","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04300","citing_title":"Argus-Retriever: Vision-LLM Late-Interaction Retrieval with Region-Aware Query-Conditioned MoE for Visual Document Retrieval","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00610","citing_title":"MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation","ref_index":43,"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":25,"is_internal_anchor":false},{"citing_arxiv_id":"2508.01959","citing_title":"SitEmb-v1.5: Improved Context-Aware Dense Retrieval for Semantic Association and Long Story Comprehension","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20354","citing_title":"EmbeddingGemma: Powerful and Lightweight Text Representations","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2405.17428","citing_title":"NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00353","citing_title":"Negative Data Mining for Contrastive Learning in Dense Retrieval at IKEA.com","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11092","citing_title":"ARHN: Answer-Centric Relabeling of Hard Negatives with Open-Source LLMs for Dense Retrieval","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08649","citing_title":"PRAGMA: Revolut Foundation Model","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07249","citing_title":"MLAIRE: Multilingual Language-Aware Information Retrieval Evaluation Protocal","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08649","citing_title":"PRAGMA: Revolut Foundation Model","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15484","citing_title":"vstash: Local-First Hybrid Retrieval with Adaptive Fusion for LLM Agents","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16576","citing_title":"On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G","json":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G.json","graph_json":"https://pith.science/api/pith-number/OFLWBQIJK6CR44E5KACTURL34G/graph.json","events_json":"https://pith.science/api/pith-number/OFLWBQIJK6CR44E5KACTURL34G/events.json","paper":"https://pith.science/paper/OFLWBQIJ"},"agent_actions":{"view_html":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G","download_json":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G.json","view_paper":"https://pith.science/paper/OFLWBQIJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.15831&json=true","fetch_graph":"https://pith.science/api/pith-number/OFLWBQIJK6CR44E5KACTURL34G/graph.json","fetch_events":"https://pith.science/api/pith-number/OFLWBQIJK6CR44E5KACTURL34G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G/action/storage_attestation","attest_author":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G/action/author_attestation","sign_citation":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G/action/citation_signature","submit_replication":"https://pith.science/pith/OFLWBQIJK6CR44E5KACTURL34G/action/replication_record"}},"created_at":"2026-07-05T10:10:42.084905+00:00","updated_at":"2026-07-05T10:10:42.084905+00:00"}