{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DOC2W33DYIXD2HMVKQ2JH6URI4","short_pith_number":"pith:DOC2W33D","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"},"canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","source":{"kind":"arxiv","id":"2604.21511","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.21511","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2604.21511v2","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.21511","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"DOC2W33DYIXD","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"DOC2W33DYIXD2HMV","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"DOC2W33D","created_at":"2026-06-02T02:04:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DOC2W33DYIXD2HMVKQ2JH6URI4","target":"record","payload":{"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"},"canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","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"},"source_kind":"arxiv","source_id":"2604.21511","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UYxbsdHrz/K9Z+cSzXWN2PgLruiDz8M4UhY3c0eAj24eusq4LHrmshd1+RZnkOTb1wiKFUBViGkewVlr0v1JDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T20:55:50.389449Z"},"content_sha256":"0d0a2c895414da0ddcbc714fc98bef840661e71084d879fe671610f2100d6a48","schema_version":"1.0","event_id":"sha256:0d0a2c895414da0ddcbc714fc98bef840661e71084d879fe671610f2100d6a48"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DOC2W33DYIXD2HMVKQ2JH6URI4","target":"graph","payload":{"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"},"verdict_id":"88771eaa-bb96-4963-9333-6be14991a1b1"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HP5yCc4lG0YLo+Y4TpN8/LZOGivvqUoICTdllRvkdsTLZVT+6mx7I9gc1mlKeopTAAiOmhMGQuNh9P/u0aEsBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T20:55:50.390330Z"},"content_sha256":"2e907cbade080cf5a6f294ebd8955a3d1313877ae5b9b126aff65c053cfc648d","schema_version":"1.0","event_id":"sha256:2e907cbade080cf5a6f294ebd8955a3d1313877ae5b9b126aff65c053cfc648d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/bundle.json","state_url":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-08T20:55:50Z","links":{"resolver":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4","bundle":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/bundle.json","state":"https://pith.science/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DOC2W33DYIXD2HMVKQ2JH6URI4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DOC2W33DYIXD2HMVKQ2JH6URI4","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ffae18e1b962fb77c3230abd25ca467a39cec0e27ba076f53d55e1817feea78d","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-04-23T10:13:21Z","title_canon_sha256":"7809308bf8450041565e2d0a7719fe8b061d5246c5ff381f71c54341c0780830"},"schema_version":"1.0","source":{"id":"2604.21511","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.21511","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2604.21511v2","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.21511","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"DOC2W33DYIXD","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"DOC2W33DYIXD2HMV","created_at":"2026-06-02T02:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"DOC2W33D","created_at":"2026-06-02T02:04:17Z"}],"graph_snapshots":[{"event_id":"sha256:2e907cbade080cf5a6f294ebd8955a3d1313877ae5b9b126aff65c053cfc648d","target":"graph","created_at":"2026-06-02T02:04:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency."}],"snapshot_sha256":"e4bf3c4da65a7997001e3d16f869f0941c1e86bb6c883182365d10d344cc2bd4"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2604.21511/integrity.json","findings":[],"snapshot_sha256":"91559a2c424ad4746806fa56964f845e6a8d759b1e0ac150f6a3077573b0fbbb","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Basile Van Cooten, Benjamin Piwowarski, Laure Soulier, Mathias Vast, Yuxuan Zong","cross_cats":["cs.CL"],"headline":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-04-23T10:13:21Z","title":"From Tokens to Concepts: Leveraging SAE for SPLADE"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.21511","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-09T20:26:30.758694Z","id":"88771eaa-bb96-4963-9333-6be14991a1b1","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"Replacing the token vocabulary in SPLADE with semantic concepts from Sparse Auto-Encoders yields comparable retrieval performance with improved efficiency.","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.","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."}},"verdict_id":"88771eaa-bb96-4963-9333-6be14991a1b1"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0d0a2c895414da0ddcbc714fc98bef840661e71084d879fe671610f2100d6a48","target":"record","created_at":"2026-06-02T02:04:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"ffae18e1b962fb77c3230abd25ca467a39cec0e27ba076f53d55e1817feea78d","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-04-23T10:13:21Z","title_canon_sha256":"7809308bf8450041565e2d0a7719fe8b061d5246c5ff381f71c54341c0780830"},"schema_version":"1.0","source":{"id":"2604.21511","kind":"arxiv","version":2}},"canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1b85ab6f63c22e3d1d95543493fa914733aa2d06286a87aea8c6089003d1160a","first_computed_at":"2026-06-02T02:04:17.954944Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:17.954944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w6k0FyzL9xwYD3GRu9E34AnN/+pzId6PlYu/Xh79ewTy8HGU3aviG0atoyLcoTpdcNT6U4Ovx8icVgCT15HLDQ==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:17.955363Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.21511","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0d0a2c895414da0ddcbc714fc98bef840661e71084d879fe671610f2100d6a48","sha256:2e907cbade080cf5a6f294ebd8955a3d1313877ae5b9b126aff65c053cfc648d"],"state_sha256":"f5e7a0396dafcc100aa6849ccafd53a6e252f1dc22e70be88014ccfd36b5eb36"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QX0zWQFk/ZeB2xepxg8X3S5ZKFGp9td6CcLaXYfkGC1FmlP/Eeliox8HeSkmIYjpKF7K67qlnDSyJ7ySFYcBDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-08T20:55:50.394729Z","bundle_sha256":"57942beb00d99638009fc669814a909953d4994754f69a49ddd0872f2283720f"}}