{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FWO4KRDCCTYKWTKKZAUS2NTSCL","short_pith_number":"pith:FWO4KRDC","canonical_record":{"source":{"id":"2406.04165","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T15:22:33Z","cross_cats_sorted":[],"title_canon_sha256":"15e09fe65d7f6a8833be1764ffae1acbe613eeabd6cc7fb90288780de88b99af","abstract_canon_sha256":"0bbdaae3496feeadc1a783b9805238babe296651f97bcb013cd8439866ad6222"},"schema_version":"1.0"},"canonical_sha256":"2d9dc5446214f0ab4d4ac8292d367212d594e5e9c699ecbc48dfe2b7da587327","source":{"kind":"arxiv","id":"2406.04165","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04165","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04165v2","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04165","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_12","alias_value":"FWO4KRDCCTYK","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_16","alias_value":"FWO4KRDCCTYKWTKK","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_8","alias_value":"FWO4KRDC","created_at":"2026-07-05T09:38:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FWO4KRDCCTYKWTKKZAUS2NTSCL","target":"record","payload":{"canonical_record":{"source":{"id":"2406.04165","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T15:22:33Z","cross_cats_sorted":[],"title_canon_sha256":"15e09fe65d7f6a8833be1764ffae1acbe613eeabd6cc7fb90288780de88b99af","abstract_canon_sha256":"0bbdaae3496feeadc1a783b9805238babe296651f97bcb013cd8439866ad6222"},"schema_version":"1.0"},"canonical_sha256":"2d9dc5446214f0ab4d4ac8292d367212d594e5e9c699ecbc48dfe2b7da587327","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:26.900668Z","signature_b64":"L39PqMy2R70AzY+DMe7XWEmTbSi42o80GgO3poKbAMx6MuCNqxuI+wgr5sKCwVNZ3uQqi/jAiIIY4f3CLZ1HBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d9dc5446214f0ab4d4ac8292d367212d594e5e9c699ecbc48dfe2b7da587327","last_reissued_at":"2026-07-05T09:38:26.900242Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:26.900242Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.04165","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":"2026-07-05T09:38:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cLsiY5N9QGPdaQvTglKycptmY3igI7ruIh/OlmX+jyOcJWiAB4VgCZH1lYMne3ocJBlUPkYsY+nT8/w5SF0ZAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T13:54:45.118468Z"},"content_sha256":"643e5383f80cf579ae5aed7b04f30be2edc13c4665dcb6c7f12e7d327c36f1a0","schema_version":"1.0","event_id":"sha256:643e5383f80cf579ae5aed7b04f30be2edc13c4665dcb6c7f12e7d327c36f1a0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FWO4KRDCCTYKWTKKZAUS2NTSCL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Repurposing Language Models into Embedding Models: Finding the Compute-Optimal Recipe","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Albert Q. Jiang, Alicja Ziarko, Bartosz Piotrowski, Mateja Jamnik, Piotr Mi{\\l}o\\'s, Wenda Li","submitted_at":"2024-06-06T15:22:33Z","abstract_excerpt":"Text embeddings are essential for many tasks, such as document retrieval, clustering, and semantic similarity assessment. In this paper, we study how to contrastively train text embedding models in a compute-optimal fashion, given a suite of pre-trained decoder-only language models. Our innovation is an algorithm that produces optimal configurations of model sizes, data quantities, and fine-tuning methods for text-embedding models at different computational budget levels. The resulting recipe, which we obtain through extensive experiments, can be used by practitioners to make informed design c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04165","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/2406.04165/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:38:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7WtZO+MmQsfOzvFq7C2wVhrDPx0cz44gUwWwPOWtQ4Uo70pKJIQjOasK/c/DPRQ2X7nHpt6ThoHKIuNHhqhlAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T13:54:45.118838Z"},"content_sha256":"44ceb4ab448a3bfef04b453b3509ffc08ffd6cfa9f8890b78f4964dfce91f490","schema_version":"1.0","event_id":"sha256:44ceb4ab448a3bfef04b453b3509ffc08ffd6cfa9f8890b78f4964dfce91f490"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/bundle.json","state_url":"https://pith.science/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/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":"2026-07-17T13:54:45Z","links":{"resolver":"https://pith.science/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL","bundle":"https://pith.science/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/bundle.json","state":"https://pith.science/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FWO4KRDCCTYKWTKKZAUS2NTSCL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FWO4KRDCCTYKWTKKZAUS2NTSCL","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":"0bbdaae3496feeadc1a783b9805238babe296651f97bcb013cd8439866ad6222","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T15:22:33Z","title_canon_sha256":"15e09fe65d7f6a8833be1764ffae1acbe613eeabd6cc7fb90288780de88b99af"},"schema_version":"1.0","source":{"id":"2406.04165","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04165","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04165v2","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04165","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_12","alias_value":"FWO4KRDCCTYK","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_16","alias_value":"FWO4KRDCCTYKWTKK","created_at":"2026-07-05T09:38:26Z"},{"alias_kind":"pith_short_8","alias_value":"FWO4KRDC","created_at":"2026-07-05T09:38:26Z"}],"graph_snapshots":[{"event_id":"sha256:44ceb4ab448a3bfef04b453b3509ffc08ffd6cfa9f8890b78f4964dfce91f490","target":"graph","created_at":"2026-07-05T09:38:26Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.04165/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Text embeddings are essential for many tasks, such as document retrieval, clustering, and semantic similarity assessment. In this paper, we study how to contrastively train text embedding models in a compute-optimal fashion, given a suite of pre-trained decoder-only language models. Our innovation is an algorithm that produces optimal configurations of model sizes, data quantities, and fine-tuning methods for text-embedding models at different computational budget levels. The resulting recipe, which we obtain through extensive experiments, can be used by practitioners to make informed design c","authors_text":"Albert Q. Jiang, Alicja Ziarko, Bartosz Piotrowski, Mateja Jamnik, Piotr Mi{\\l}o\\'s, Wenda Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T15:22:33Z","title":"Repurposing Language Models into Embedding Models: Finding the Compute-Optimal Recipe"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04165","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:643e5383f80cf579ae5aed7b04f30be2edc13c4665dcb6c7f12e7d327c36f1a0","target":"record","created_at":"2026-07-05T09:38:26Z","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":"0bbdaae3496feeadc1a783b9805238babe296651f97bcb013cd8439866ad6222","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T15:22:33Z","title_canon_sha256":"15e09fe65d7f6a8833be1764ffae1acbe613eeabd6cc7fb90288780de88b99af"},"schema_version":"1.0","source":{"id":"2406.04165","kind":"arxiv","version":2}},"canonical_sha256":"2d9dc5446214f0ab4d4ac8292d367212d594e5e9c699ecbc48dfe2b7da587327","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2d9dc5446214f0ab4d4ac8292d367212d594e5e9c699ecbc48dfe2b7da587327","first_computed_at":"2026-07-05T09:38:26.900242Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:38:26.900242Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L39PqMy2R70AzY+DMe7XWEmTbSi42o80GgO3poKbAMx6MuCNqxuI+wgr5sKCwVNZ3uQqi/jAiIIY4f3CLZ1HBg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:38:26.900668Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.04165","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:643e5383f80cf579ae5aed7b04f30be2edc13c4665dcb6c7f12e7d327c36f1a0","sha256:44ceb4ab448a3bfef04b453b3509ffc08ffd6cfa9f8890b78f4964dfce91f490"],"state_sha256":"c699b2e1953e068c84d86c8c19268ed27d1ff5342b43720d87db6b05364421c0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wq2CHKhCSeSNDMuOzT74nIPJJPi1XDEGZmpI3E3TYTLa09wQ90/jLJeC6bqvJJMV9sFQpkvt6ldUG2g9HCO1Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-17T13:54:45.120936Z","bundle_sha256":"5af20afcdb3b549dcd5ff30b116151ff8dc3469d4a07a1507bb5f76196e0a58b"}}