{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DOC2W33DYIXD2HMVKQ2JH6URI4","short_pith_number":"pith:DOC2W33D","schema_version":"1.0","canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","source":{"kind":"arxiv","id":"2604.21511","version":2},"attestation_state":"computed","paper":{"title":"From Tokens to Concepts: Leveraging SAE for SPLADE","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency.","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Basile Van Cooten, Benjamin Piwowarski, Laure Soulier, Mathias Vast, Yuxuan Zong","submitted_at":"2026-04-23T10:13:21Z","abstract_excerpt":"Learned Sparse IR models, such as SPLADE, offer an excellent efficiency-effectiveness tradeoff. However, they rely on the underlying backbone vocabulary, which might hinder performance (polysemicity and synonymy) and pose a challenge for multi-lingual and multi-modal usages. To solve this limitation, we propose to replace the backbone vocabulary with a latent space of semantic concepts learned using Sparse Auto-Encoders (SAE). Throughout this paper, we study the compatibility of these 2 concepts, explore training approaches, and analyze the differences between our SAE-SPLADE model and traditio"},"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":"2604.21511","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-04-23T10:13:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"7809308bf8450041565e2d0a7719fe8b061d5246c5ff381f71c54341c0780830","abstract_canon_sha256":"ffae18e1b962fb77c3230abd25ca467a39cec0e27ba076f53d55e1817feea78d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:17.955363Z","signature_b64":"w6k0FyzL9xwYD3GRu9E34AnN/+pzId6PlYu/Xh79ewTy8HGU3aviG0atoyLcoTpdcNT6U4Ovx8icVgCT15HLDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","last_reissued_at":"2026-06-02T02:04:17.954944Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:17.954944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Tokens to Concepts: Leveraging SAE for SPLADE","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency.","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Basile Van Cooten, Benjamin Piwowarski, Laure Soulier, Mathias Vast, Yuxuan Zong","submitted_at":"2026-04-23T10:13:21Z","abstract_excerpt":"Learned Sparse IR models, such as SPLADE, offer an excellent efficiency-effectiveness tradeoff. However, they rely on the underlying backbone vocabulary, which might hinder performance (polysemicity and synonymy) and pose a challenge for multi-lingual and multi-modal usages. To solve this limitation, we propose to replace the backbone vocabulary with a latent space of semantic concepts learned using Sparse Auto-Encoders (SAE). Throughout this paper, we study the compatibility of these 2 concepts, explore training approaches, and analyze the differences between our SAE-SPLADE model and traditio"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experiments demonstrate that SAE-SPLADE achieves retrieval performance comparable to SPLADE on both in-domain and out-of-domain tasks while offering improved efficiency.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the SAE-learned latent concepts form a drop-in replacement for the original token vocabulary without losing retrieval-critical information and that the two training regimes remain compatible.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SAE-SPLADE substitutes SPLADE's backbone vocabulary with SAE-derived semantic concepts and matches standard SPLADE performance with better efficiency on in- and out-of-domain tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e4bf3c4da65a7997001e3d16f869f0941c1e86bb6c883182365d10d344cc2bd4"},"source":{"id":"2604.21511","kind":"arxiv","version":2},"verdict":{"id":"88771eaa-bb96-4963-9333-6be14991a1b1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T20:26:30.758694Z","strongest_claim":"Our experiments demonstrate that SAE-SPLADE achieves retrieval performance comparable to SPLADE on both in-domain and out-of-domain tasks while offering improved efficiency.","one_line_summary":"SAE-SPLADE substitutes SPLADE's backbone vocabulary with SAE-derived semantic concepts and matches standard SPLADE performance with better efficiency on in- and out-of-domain tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the SAE-learned latent concepts form a drop-in replacement for the original token vocabulary without losing retrieval-critical information and that the two training regimes remain compatible.","pith_extraction_headline":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.21511/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T12:39:54.431790Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-20T00:56:40.743827Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"91559a2c424ad4746806fa56964f845e6a8d759b1e0ac150f6a3077573b0fbbb"},"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":"2604.21511","created_at":"2026-06-02T02:04:17.954999+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.21511v2","created_at":"2026-06-02T02:04:17.954999+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.21511","created_at":"2026-06-02T02:04:17.954999+00:00"},{"alias_kind":"pith_short_12","alias_value":"DOC2W33DYIXD","created_at":"2026-06-02T02:04:17.954999+00:00"},{"alias_kind":"pith_short_16","alias_value":"DOC2W33DYIXD2HMV","created_at":"2026-06-02T02:04:17.954999+00:00"},{"alias_kind":"pith_short_8","alias_value":"DOC2W33D","created_at":"2026-06-02T02:04:17.954999+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/DOC2W33DYIXD2HMVKQ2JH6URI4","json":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4.json","graph_json":"https://pith.science/api/pith-number/DOC2W33DYIXD2HMVKQ2JH6URI4/graph.json","events_json":"https://pith.science/api/pith-number/DOC2W33DYIXD2HMVKQ2JH6URI4/events.json","paper":"https://pith.science/paper/DOC2W33D"},"agent_actions":{"view_html":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4","download_json":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4.json","view_paper":"https://pith.science/paper/DOC2W33D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.21511&json=true","fetch_graph":"https://pith.science/api/pith-number/DOC2W33DYIXD2HMVKQ2JH6URI4/graph.json","fetch_events":"https://pith.science/api/pith-number/DOC2W33DYIXD2HMVKQ2JH6URI4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/action/storage_attestation","attest_author":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/action/author_attestation","sign_citation":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/action/citation_signature","submit_replication":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/action/replication_record"}},"created_at":"2026-06-02T02:04:17.954999+00:00","updated_at":"2026-06-02T02:04:17.954999+00:00"}