{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SKFFHLVSU5LAZYVPCVCBJ7V6EF","short_pith_number":"pith:SKFFHLVS","canonical_record":{"source":{"id":"2410.01296","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T07:42:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c4bfa9a3d8a4f6362876efdc047d9a1dbf747c5783ee3ef0eb771ebc0762d363","abstract_canon_sha256":"f5ec3c031d07dca18b21f8fa8a2f48fa0780218eca8798595e969418d63d830f"},"schema_version":"1.0"},"canonical_sha256":"928a53aeb2a7560ce2af154414febe2174be36d3f5dd890c1af48553d3a45d1a","source":{"kind":"arxiv","id":"2410.01296","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.01296","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"arxiv_version","alias_value":"2410.01296v1","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01296","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_12","alias_value":"SKFFHLVSU5LA","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_16","alias_value":"SKFFHLVSU5LAZYVP","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_8","alias_value":"SKFFHLVS","created_at":"2026-07-05T09:14:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SKFFHLVSU5LAZYVPCVCBJ7V6EF","target":"record","payload":{"canonical_record":{"source":{"id":"2410.01296","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T07:42:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c4bfa9a3d8a4f6362876efdc047d9a1dbf747c5783ee3ef0eb771ebc0762d363","abstract_canon_sha256":"f5ec3c031d07dca18b21f8fa8a2f48fa0780218eca8798595e969418d63d830f"},"schema_version":"1.0"},"canonical_sha256":"928a53aeb2a7560ce2af154414febe2174be36d3f5dd890c1af48553d3a45d1a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:14:45.894477Z","signature_b64":"ZO21kE+kE5ZmqXpyWAeBQW1ipsq+8KFZvwZvuKYVkvgIxY833e/QvhkELWPnmuuEcCttFvENj0uviLCrFK93Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"928a53aeb2a7560ce2af154414febe2174be36d3f5dd890c1af48553d3a45d1a","last_reissued_at":"2026-07-05T09:14:45.894023Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:14:45.894023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.01296","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:14:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d4qvGd/LXWqfscR2Gb/2XQgTrLmi27UhgGS2BvgqAAXWCrs6W44gdDybnlQN0IrMakZsl7DPgUfWsOpNQ0wMCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:38.088792Z"},"content_sha256":"d0de68d5bcc6254ffe17041801c209b406f54081a35dcaed1db08ba00aa5363e","schema_version":"1.0","event_id":"sha256:d0de68d5bcc6254ffe17041801c209b406f54081a35dcaed1db08ba00aa5363e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SKFFHLVSU5LAZYVPCVCBJ7V6EF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Speculative Coreset Selection for Task-Specific Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chao Shen, Juan Zhai, Shiqing Ma, Tianlin Li, Weipeng Jiang, Xiaoyu Zhang, Yang Liu","submitted_at":"2024-10-02T07:42:25Z","abstract_excerpt":"Task-specific fine-tuning is essential for the deployment of large language models (LLMs), but it requires significant computational resources and time. Existing solutions have proposed coreset selection methods to improve data efficiency and reduce model training overhead, but they still have limitations: 1) Overlooking valuable samples at high pruning rates, which degrades the coreset's performance. 2) Requiring high time overhead during coreset selection to fine-tune and evaluate the target LLM. In this paper, we introduce STAFF, a speculative coreset selection method. STAFF leverages a sma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01296","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/2410.01296/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":"2026-07-05T09:14:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"63Dju2wPTujpxWn7bqM2TI61ux6Ph/X4qdRmq2BBHH5OkIQ8yiGu2N0IgJ0110THQRLUIeUdzh4e7ioIHBdhBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:38.089285Z"},"content_sha256":"f737a239cb8963e2e0b995f9907d2c410a81d129ded63e8e49ca4a09aee2219a","schema_version":"1.0","event_id":"sha256:f737a239cb8963e2e0b995f9907d2c410a81d129ded63e8e49ca4a09aee2219a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/bundle.json","state_url":"https://pith.science/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/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":"2026-08-08T13:13:38Z","links":{"resolver":"https://pith.science/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF","bundle":"https://pith.science/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/bundle.json","state":"https://pith.science/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SKFFHLVSU5LAZYVPCVCBJ7V6EF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SKFFHLVSU5LAZYVPCVCBJ7V6EF","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":"f5ec3c031d07dca18b21f8fa8a2f48fa0780218eca8798595e969418d63d830f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T07:42:25Z","title_canon_sha256":"c4bfa9a3d8a4f6362876efdc047d9a1dbf747c5783ee3ef0eb771ebc0762d363"},"schema_version":"1.0","source":{"id":"2410.01296","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.01296","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"arxiv_version","alias_value":"2410.01296v1","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01296","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_12","alias_value":"SKFFHLVSU5LA","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_16","alias_value":"SKFFHLVSU5LAZYVP","created_at":"2026-07-05T09:14:45Z"},{"alias_kind":"pith_short_8","alias_value":"SKFFHLVS","created_at":"2026-07-05T09:14:45Z"}],"graph_snapshots":[{"event_id":"sha256:f737a239cb8963e2e0b995f9907d2c410a81d129ded63e8e49ca4a09aee2219a","target":"graph","created_at":"2026-07-05T09:14:45Z","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/2410.01296/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Task-specific fine-tuning is essential for the deployment of large language models (LLMs), but it requires significant computational resources and time. Existing solutions have proposed coreset selection methods to improve data efficiency and reduce model training overhead, but they still have limitations: 1) Overlooking valuable samples at high pruning rates, which degrades the coreset's performance. 2) Requiring high time overhead during coreset selection to fine-tune and evaluate the target LLM. In this paper, we introduce STAFF, a speculative coreset selection method. STAFF leverages a sma","authors_text":"Chao Shen, Juan Zhai, Shiqing Ma, Tianlin Li, Weipeng Jiang, Xiaoyu Zhang, Yang Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T07:42:25Z","title":"Speculative Coreset Selection for Task-Specific Fine-tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01296","kind":"arxiv","version":1},"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:d0de68d5bcc6254ffe17041801c209b406f54081a35dcaed1db08ba00aa5363e","target":"record","created_at":"2026-07-05T09:14:45Z","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":"f5ec3c031d07dca18b21f8fa8a2f48fa0780218eca8798595e969418d63d830f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T07:42:25Z","title_canon_sha256":"c4bfa9a3d8a4f6362876efdc047d9a1dbf747c5783ee3ef0eb771ebc0762d363"},"schema_version":"1.0","source":{"id":"2410.01296","kind":"arxiv","version":1}},"canonical_sha256":"928a53aeb2a7560ce2af154414febe2174be36d3f5dd890c1af48553d3a45d1a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"928a53aeb2a7560ce2af154414febe2174be36d3f5dd890c1af48553d3a45d1a","first_computed_at":"2026-07-05T09:14:45.894023Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:14:45.894023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZO21kE+kE5ZmqXpyWAeBQW1ipsq+8KFZvwZvuKYVkvgIxY833e/QvhkELWPnmuuEcCttFvENj0uviLCrFK93Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:14:45.894477Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.01296","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d0de68d5bcc6254ffe17041801c209b406f54081a35dcaed1db08ba00aa5363e","sha256:f737a239cb8963e2e0b995f9907d2c410a81d129ded63e8e49ca4a09aee2219a"],"state_sha256":"08f6b375f1939f5de37304775f7076ee44da8356098db40774f586c196f27f63"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sgsvnu33u0/8eEKWY9M+0JqOxVvS4Lw8rLlu7i3VF0wYztoFn8TtODg3brcy7m+Vj4CyiDJX43oXr4BOdjYpDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:13:38.094617Z","bundle_sha256":"677f057e7db2075d1ad3829981fae8999a76925caa3e3f4c164a508eafffc121"}}