{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ITDHYBZM4ZBVHEMQV7E23MBRFO","short_pith_number":"pith:ITDHYBZM","schema_version":"1.0","canonical_sha256":"44c67c072ce643539190afc9adb0312ba84ebd062270554715954dfcd73570df","source":{"kind":"arxiv","id":"2502.02788","version":1},"attestation_state":"computed","paper":{"title":"Inducing Diversity in Differentiable Search Indexing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Abhijeet Phatak, Chittaranjan Tripathy, Jayant Sachdev, Sean D Rosario, Swati Kirti","submitted_at":"2025-02-05T00:21:17Z","abstract_excerpt":"Differentiable Search Indexing (DSI) is a recent paradigm for information retrieval which uses a transformer-based neural network architecture as the document index to simplify the retrieval process. A differentiable index has many advantages enabling modifications, updates or extensions to the index. In this work, we explore balancing relevance and novel information content (diversity) for training DSI systems inspired by Maximal Marginal Relevance (MMR), and show the benefits of our approach over the naive DSI training. We present quantitative and qualitative evaluations of relevance and div"},"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":"2502.02788","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-02-05T00:21:17Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"51bdf2866aebab110ca3a65fb64173e9889d8448d3b102977f1fe290a85bbe94","abstract_canon_sha256":"4fa29a98bdd46fd1634613c9afcccd88f135dd38e5485106517309c0347badcd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:09:55.179328Z","signature_b64":"nbS9kXYy2VuEy935zOu4Pi8K1izF6+h+FOrnWqM2i4EBLP3lpR8y1g/SVDr9u4i7AUyCNjH7BNVSHjr6sYwXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44c67c072ce643539190afc9adb0312ba84ebd062270554715954dfcd73570df","last_reissued_at":"2026-07-05T10:09:55.178843Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:09:55.178843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inducing Diversity in Differentiable Search Indexing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Abhijeet Phatak, Chittaranjan Tripathy, Jayant Sachdev, Sean D Rosario, Swati Kirti","submitted_at":"2025-02-05T00:21:17Z","abstract_excerpt":"Differentiable Search Indexing (DSI) is a recent paradigm for information retrieval which uses a transformer-based neural network architecture as the document index to simplify the retrieval process. A differentiable index has many advantages enabling modifications, updates or extensions to the index. In this work, we explore balancing relevance and novel information content (diversity) for training DSI systems inspired by Maximal Marginal Relevance (MMR), and show the benefits of our approach over the naive DSI training. We present quantitative and qualitative evaluations of relevance and div"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.02788","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/2502.02788/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":"2502.02788","created_at":"2026-07-05T10:09:55.178905+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.02788v1","created_at":"2026-07-05T10:09:55.178905+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.02788","created_at":"2026-07-05T10:09:55.178905+00:00"},{"alias_kind":"pith_short_12","alias_value":"ITDHYBZM4ZBV","created_at":"2026-07-05T10:09:55.178905+00:00"},{"alias_kind":"pith_short_16","alias_value":"ITDHYBZM4ZBVHEMQ","created_at":"2026-07-05T10:09:55.178905+00:00"},{"alias_kind":"pith_short_8","alias_value":"ITDHYBZM","created_at":"2026-07-05T10:09:55.178905+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/ITDHYBZM4ZBVHEMQV7E23MBRFO","json":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO.json","graph_json":"https://pith.science/api/pith-number/ITDHYBZM4ZBVHEMQV7E23MBRFO/graph.json","events_json":"https://pith.science/api/pith-number/ITDHYBZM4ZBVHEMQV7E23MBRFO/events.json","paper":"https://pith.science/paper/ITDHYBZM"},"agent_actions":{"view_html":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO","download_json":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO.json","view_paper":"https://pith.science/paper/ITDHYBZM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.02788&json=true","fetch_graph":"https://pith.science/api/pith-number/ITDHYBZM4ZBVHEMQV7E23MBRFO/graph.json","fetch_events":"https://pith.science/api/pith-number/ITDHYBZM4ZBVHEMQV7E23MBRFO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO/action/storage_attestation","attest_author":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO/action/author_attestation","sign_citation":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO/action/citation_signature","submit_replication":"https://pith.science/pith/ITDHYBZM4ZBVHEMQV7E23MBRFO/action/replication_record"}},"created_at":"2026-07-05T10:09:55.178905+00:00","updated_at":"2026-07-05T10:09:55.178905+00:00"}