{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:R6OZW2ZMGK5UJYFNVHE7HBO6XS","short_pith_number":"pith:R6OZW2ZM","canonical_record":{"source":{"id":"2408.10613","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-20T07:48:19Z","cross_cats_sorted":[],"title_canon_sha256":"820fadbf4bb3b0af19443174bacb9c7ec62988f8374d0314800d07112bb94144","abstract_canon_sha256":"754535cf0913c73d1bef81760c557660be6ba07b52e3f2d86d5ce74a7115d394"},"schema_version":"1.0"},"canonical_sha256":"8f9d9b6b2c32bb44e0ada9c9f385debc8c54ed96df1c803013015dcdb91d0af1","source":{"kind":"arxiv","id":"2408.10613","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.10613","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"arxiv_version","alias_value":"2408.10613v3","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.10613","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_12","alias_value":"R6OZW2ZMGK5U","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_16","alias_value":"R6OZW2ZMGK5UJYFN","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_8","alias_value":"R6OZW2ZM","created_at":"2026-07-05T11:02:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:R6OZW2ZMGK5UJYFNVHE7HBO6XS","target":"record","payload":{"canonical_record":{"source":{"id":"2408.10613","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-20T07:48:19Z","cross_cats_sorted":[],"title_canon_sha256":"820fadbf4bb3b0af19443174bacb9c7ec62988f8374d0314800d07112bb94144","abstract_canon_sha256":"754535cf0913c73d1bef81760c557660be6ba07b52e3f2d86d5ce74a7115d394"},"schema_version":"1.0"},"canonical_sha256":"8f9d9b6b2c32bb44e0ada9c9f385debc8c54ed96df1c803013015dcdb91d0af1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:12.230906Z","signature_b64":"6h+p9uanilDHKp5+2rmOIkTNw+Em9kxVqBjDJFlkDHrPwc8H1bPS76M2OJK1uXpIKAqfCc/G4DkTo7xnL5XNCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f9d9b6b2c32bb44e0ada9c9f385debc8c54ed96df1c803013015dcdb91d0af1","last_reissued_at":"2026-07-05T11:02:12.230422Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:12.230422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.10613","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:02:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SXLM3duzXWVTS5e5lSiGOppWFFrFWbL9Nq054ctxof0zP/j71jEsn1H5mRPagq9nDrioajqClgnYwl3/YGx2Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T16:45:02.444976Z"},"content_sha256":"844b5a9ba417210217b81d9b39faa38eebdb0f849986ca5afe287ef9e8911c60","schema_version":"1.0","event_id":"sha256:844b5a9ba417210217b81d9b39faa38eebdb0f849986ca5afe287ef9e8911c60"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:R6OZW2ZMGK5UJYFNVHE7HBO6XS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Task-level Distributionally Robust Optimization for Large Language Model-based Dense Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Guangyuan Ma, Ming Zhou, Songlin Hu, Xing Wu, Yongliang Ma, Zhenpeng Su","submitted_at":"2024-08-20T07:48:19Z","abstract_excerpt":"Large Language Model-based Dense Retrieval (LLM-DR) optimizes over numerous heterogeneous fine-tuning collections from different domains. However, the discussion about its training data distribution is still minimal. Previous studies rely on empirically assigned dataset choices or sampling ratios, which inevitably lead to sub-optimal retrieval performances. In this paper, we propose a new task-level Distributionally Robust Optimization (tDRO) algorithm for LLM-DR fine-tuning, targeted at improving the universal domain generalization ability by end-to-end reweighting the data distribution of ea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10613","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/2408.10613/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":"2026-07-05T11:02:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3uPieYr9KiT0scEMS2+3iJt/wNPGzymtYo7Kk0NRm6u+xTP36ss82aItMR8hE5IUhpKx+L7SkMl0U/qQUdriAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T16:45:02.445498Z"},"content_sha256":"cf47a902f8b64144c5c9e3c117d06a985a8e24ab18412fc6cf4fee0ad316a33c","schema_version":"1.0","event_id":"sha256:cf47a902f8b64144c5c9e3c117d06a985a8e24ab18412fc6cf4fee0ad316a33c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/bundle.json","state_url":"https://pith.science/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/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":"2026-08-18T16:45:02Z","links":{"resolver":"https://pith.science/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS","bundle":"https://pith.science/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/bundle.json","state":"https://pith.science/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R6OZW2ZMGK5UJYFNVHE7HBO6XS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:R6OZW2ZMGK5UJYFNVHE7HBO6XS","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":"754535cf0913c73d1bef81760c557660be6ba07b52e3f2d86d5ce74a7115d394","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-20T07:48:19Z","title_canon_sha256":"820fadbf4bb3b0af19443174bacb9c7ec62988f8374d0314800d07112bb94144"},"schema_version":"1.0","source":{"id":"2408.10613","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.10613","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"arxiv_version","alias_value":"2408.10613v3","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.10613","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_12","alias_value":"R6OZW2ZMGK5U","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_16","alias_value":"R6OZW2ZMGK5UJYFN","created_at":"2026-07-05T11:02:12Z"},{"alias_kind":"pith_short_8","alias_value":"R6OZW2ZM","created_at":"2026-07-05T11:02:12Z"}],"graph_snapshots":[{"event_id":"sha256:cf47a902f8b64144c5c9e3c117d06a985a8e24ab18412fc6cf4fee0ad316a33c","target":"graph","created_at":"2026-07-05T11:02:12Z","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/2408.10613/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Model-based Dense Retrieval (LLM-DR) optimizes over numerous heterogeneous fine-tuning collections from different domains. However, the discussion about its training data distribution is still minimal. Previous studies rely on empirically assigned dataset choices or sampling ratios, which inevitably lead to sub-optimal retrieval performances. In this paper, we propose a new task-level Distributionally Robust Optimization (tDRO) algorithm for LLM-DR fine-tuning, targeted at improving the universal domain generalization ability by end-to-end reweighting the data distribution of ea","authors_text":"Guangyuan Ma, Ming Zhou, Songlin Hu, Xing Wu, Yongliang Ma, Zhenpeng Su","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-20T07:48:19Z","title":"Task-level Distributionally Robust Optimization for Large Language Model-based Dense Retrieval"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10613","kind":"arxiv","version":3},"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":"sha256:844b5a9ba417210217b81d9b39faa38eebdb0f849986ca5afe287ef9e8911c60","target":"record","created_at":"2026-07-05T11:02:12Z","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":"754535cf0913c73d1bef81760c557660be6ba07b52e3f2d86d5ce74a7115d394","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-20T07:48:19Z","title_canon_sha256":"820fadbf4bb3b0af19443174bacb9c7ec62988f8374d0314800d07112bb94144"},"schema_version":"1.0","source":{"id":"2408.10613","kind":"arxiv","version":3}},"canonical_sha256":"8f9d9b6b2c32bb44e0ada9c9f385debc8c54ed96df1c803013015dcdb91d0af1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8f9d9b6b2c32bb44e0ada9c9f385debc8c54ed96df1c803013015dcdb91d0af1","first_computed_at":"2026-07-05T11:02:12.230422Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:02:12.230422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6h+p9uanilDHKp5+2rmOIkTNw+Em9kxVqBjDJFlkDHrPwc8H1bPS76M2OJK1uXpIKAqfCc/G4DkTo7xnL5XNCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:02:12.230906Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.10613","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:844b5a9ba417210217b81d9b39faa38eebdb0f849986ca5afe287ef9e8911c60","sha256:cf47a902f8b64144c5c9e3c117d06a985a8e24ab18412fc6cf4fee0ad316a33c"],"state_sha256":"a688fe7b379c48b2352b52908920d21ed61f3e465ad65914fe6072732935fa30"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LfHyaBQrnyF5mgVIBllvyzSzwOVeGuWhB8SNSVjIWaG6P9upHFLrkDLrpP3TbzjhfCo+Ca60ILCihae2qrYpAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T16:45:02.450641Z","bundle_sha256":"41916e83ba2ce65896f07fab150d41c19e055b628c6a158c948e3fa424fb5717"}}