{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:KBNTB5YJJG3UTXK4OIXCSUVMKL","short_pith_number":"pith:KBNTB5YJ","schema_version":"1.0","canonical_sha256":"505b30f70949b749dd5c722e2952ac52e699f8d4e9c93ceff544d1b2bca19f26","source":{"kind":"arxiv","id":"2112.01488","version":3},"attestation_state":"computed","paper":{"title":"ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Christopher Potts, Jon Saad-Falcon, Keshav Santhanam, Matei Zaharia, Omar Khattab","submitted_at":"2021-12-02T18:38:50Z","abstract_excerpt":"Neural information retrieval (IR) has greatly advanced search and other knowledge-intensive language tasks. While many neural IR methods encode queries and documents into single-vector representations, late interaction models produce multi-vector representations at the granularity of each token and decompose relevance modeling into scalable token-level computations. This decomposition has been shown to make late interaction more effective, but it inflates the space footprint of these models by an order of magnitude. In this work, we introduce ColBERTv2, a retriever that couples an aggressive r"},"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":"2112.01488","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-12-02T18:38:50Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"008df0cf6c9d35d3d739de3c07e4338c7686e236723bf7855098d4a120b2ac17","abstract_canon_sha256":"8a331955c6f1b7b9eaf3b770d371f3c884defdf25bab54ab75514c39b0208dc3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:38:46.284177Z","signature_b64":"HeDMoldk3EHBOoVDLTQL600zUS/OqV6xdpM8w+5+CO8CW3OetIJYj5zEQu/f1596S0k6NjGTGvTkz4ZvRqX7Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"505b30f70949b749dd5c722e2952ac52e699f8d4e9c93ceff544d1b2bca19f26","last_reissued_at":"2026-07-05T04:38:46.283723Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:38:46.283723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Christopher Potts, Jon Saad-Falcon, Keshav Santhanam, Matei Zaharia, Omar Khattab","submitted_at":"2021-12-02T18:38:50Z","abstract_excerpt":"Neural information retrieval (IR) has greatly advanced search and other knowledge-intensive language tasks. While many neural IR methods encode queries and documents into single-vector representations, late interaction models produce multi-vector representations at the granularity of each token and decompose relevance modeling into scalable token-level computations. This decomposition has been shown to make late interaction more effective, but it inflates the space footprint of these models by an order of magnitude. In this work, we introduce ColBERTv2, a retriever that couples an aggressive r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.01488","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/2112.01488/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":"2112.01488","created_at":"2026-07-05T04:38:46.283782+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.01488v3","created_at":"2026-07-05T04:38:46.283782+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.01488","created_at":"2026-07-05T04:38:46.283782+00:00"},{"alias_kind":"pith_short_12","alias_value":"KBNTB5YJJG3U","created_at":"2026-07-05T04:38:46.283782+00:00"},{"alias_kind":"pith_short_16","alias_value":"KBNTB5YJJG3UTXK4","created_at":"2026-07-05T04:38:46.283782+00:00"},{"alias_kind":"pith_short_8","alias_value":"KBNTB5YJ","created_at":"2026-07-05T04:38:46.283782+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23475","citing_title":"Multi-Vector Embeddings are Provably More Expressive than Single Vector Embeddings","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22778","citing_title":"HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions","ref_index":117,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01240","citing_title":"Efficient RAG with Intent-Aware Retrieval and Semantics-Preserving Chunking","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24297","citing_title":"Benchmarking Patent Embeddings: A Multi-Task Evaluation of 22 Models Across Retrieval, Classification, and Clustering","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24764","citing_title":"Spectral Retrieval: Multi-Scale Sinc Convolution over Token Embeddings for Localized Retrieval in LLM Multi-Agent Systems","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28365","citing_title":"CAMI: Cost-Aware Agent-Guided Multi-Indexing for Semantic Retrieval","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2512.05958","citing_title":"MaxShapley: Towards Incentive-compatible Generative Search with Fair Context Attribution","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18769","citing_title":"ClusterRAG: Cluster-Based Collaborative Filtering for Personalized Retrieval-Augmented Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19874","citing_title":"TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2509.12539","citing_title":"LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2510.05038","citing_title":"Guided Query Refinement: Multimodal Hybrid Retrieval with Test-Time Optimization","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2201.10005","citing_title":"Text and Code Embeddings by Contrastive Pre-Training","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2401.18059","citing_title":"RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval","ref_index":106,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20354","citing_title":"EmbeddingGemma: Powerful and Lightweight Text Representations","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14169","citing_title":"Chronological Knowledge Retrieval: A Retrieval-Augmented Generation Approach to Construction Project Documentation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2310.03714","citing_title":"DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00702","citing_title":"Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory","ref_index":62,"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":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL","json":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL.json","graph_json":"https://pith.science/api/pith-number/KBNTB5YJJG3UTXK4OIXCSUVMKL/graph.json","events_json":"https://pith.science/api/pith-number/KBNTB5YJJG3UTXK4OIXCSUVMKL/events.json","paper":"https://pith.science/paper/KBNTB5YJ"},"agent_actions":{"view_html":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL","download_json":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL.json","view_paper":"https://pith.science/paper/KBNTB5YJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.01488&json=true","fetch_graph":"https://pith.science/api/pith-number/KBNTB5YJJG3UTXK4OIXCSUVMKL/graph.json","fetch_events":"https://pith.science/api/pith-number/KBNTB5YJJG3UTXK4OIXCSUVMKL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL/action/storage_attestation","attest_author":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL/action/author_attestation","sign_citation":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL/action/citation_signature","submit_replication":"https://pith.science/pith/KBNTB5YJJG3UTXK4OIXCSUVMKL/action/replication_record"}},"created_at":"2026-07-05T04:38:46.283782+00:00","updated_at":"2026-07-05T04:38:46.283782+00:00"}