{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:35254I5XIPRXMRHLZR2I6LI46A","short_pith_number":"pith:35254I5X","canonical_record":{"source":{"id":"2603.12478","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-12T21:54:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4ad27a4050331b6b74659e510b5d9e9a78fff4d54b4bfe53787e75100a3acdee","abstract_canon_sha256":"64f0e75e59b9bc92fba4e1b9c6408393b6ae70d0046af2932782ffdc0f539006"},"schema_version":"1.0"},"canonical_sha256":"df75de23b743e37644ebcc748f2d1cf035066eb1335d5694a0f5f1a0662b261c","source":{"kind":"arxiv","id":"2603.12478","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.12478","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"arxiv_version","alias_value":"2603.12478v2","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.12478","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_12","alias_value":"35254I5XIPRX","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_16","alias_value":"35254I5XIPRXMRHL","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_8","alias_value":"35254I5X","created_at":"2026-07-10T01:18:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:35254I5XIPRXMRHLZR2I6LI46A","target":"record","payload":{"canonical_record":{"source":{"id":"2603.12478","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-12T21:54:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4ad27a4050331b6b74659e510b5d9e9a78fff4d54b4bfe53787e75100a3acdee","abstract_canon_sha256":"64f0e75e59b9bc92fba4e1b9c6408393b6ae70d0046af2932782ffdc0f539006"},"schema_version":"1.0"},"canonical_sha256":"df75de23b743e37644ebcc748f2d1cf035066eb1335d5694a0f5f1a0662b261c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-10T01:18:58.868121Z","signature_b64":"vCKVaA4rgTDqVLs4GvgNSiMAHx8iIZcxBfkig3sCwn3L70zHfEt+IstLKKxnhNOJiXwNI7NUJafHD5l7v8oUDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df75de23b743e37644ebcc748f2d1cf035066eb1335d5694a0f5f1a0662b261c","last_reissued_at":"2026-07-10T01:18:58.867623Z","signature_status":"signed_v1","first_computed_at":"2026-07-10T01:18:58.867623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.12478","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":"2026-07-10T01:18:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wtLCz3IJKeRvp6ZuIoIfQBe/D1Sv6+WI6vZAVV2IYzjSpkXIP2B7QcWwmRNuqKpyX6dSfKhrks+PgAVeNH2xAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T12:02:13.311684Z"},"content_sha256":"3b469953d28feb1fead1effc1d4db6fa9626f68fa785b52107b1a2bd33b68923","schema_version":"1.0","event_id":"sha256:3b469953d28feb1fead1effc1d4db6fa9626f68fa785b52107b1a2bd33b68923"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:35254I5XIPRXMRHLZR2I6LI46A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Less Data, Faster Convergence: Goal-Driven Data Optimization for Multimodal Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Hai Ci, Haozhe Zhao, Rujie Wu, Yizhou Wang","submitted_at":"2026-03-12T21:54:50Z","abstract_excerpt":"Multimodal instruction tuning is often compute-inefficient because training budgets are spread across large mixed image-video pools whose utility is highly uneven. We present Goal-Driven Data Optimization (GDO), a framework that computes six sample descriptors for each candidate and constructs optimized 1$\\times$ training subsets for different goals. Under a fixed one-epoch Qwen3-VL-8B-Instruct training and evaluation recipe on 8 H20 GPUs, GDO uses far fewer training samples than the Uni-10x baseline while converging faster and achieving higher accuracy. Relative to the fixed 512k-sample Uni-1"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.12478","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/2603.12478/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":"2026-07-10T01:18:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VhxKpIZlMQv3pKJDcKDnrWpc46qazArBFE+p6X6lCLbyCpWThlBqffeUh2assRMk+Hz9VOTMIMaV7w+nACh2BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T12:02:13.311974Z"},"content_sha256":"6a9a504b7106f8296944630f1d496df1c0337cb67b90386de5a737486785855f","schema_version":"1.0","event_id":"sha256:6a9a504b7106f8296944630f1d496df1c0337cb67b90386de5a737486785855f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/35254I5XIPRXMRHLZR2I6LI46A/bundle.json","state_url":"https://pith.science/pith/35254I5XIPRXMRHLZR2I6LI46A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/35254I5XIPRXMRHLZR2I6LI46A/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":"2026-08-01T12:02:13Z","links":{"resolver":"https://pith.science/pith/35254I5XIPRXMRHLZR2I6LI46A","bundle":"https://pith.science/pith/35254I5XIPRXMRHLZR2I6LI46A/bundle.json","state":"https://pith.science/pith/35254I5XIPRXMRHLZR2I6LI46A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/35254I5XIPRXMRHLZR2I6LI46A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:35254I5XIPRXMRHLZR2I6LI46A","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":"64f0e75e59b9bc92fba4e1b9c6408393b6ae70d0046af2932782ffdc0f539006","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-12T21:54:50Z","title_canon_sha256":"4ad27a4050331b6b74659e510b5d9e9a78fff4d54b4bfe53787e75100a3acdee"},"schema_version":"1.0","source":{"id":"2603.12478","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.12478","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"arxiv_version","alias_value":"2603.12478v2","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.12478","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_12","alias_value":"35254I5XIPRX","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_16","alias_value":"35254I5XIPRXMRHL","created_at":"2026-07-10T01:18:58Z"},{"alias_kind":"pith_short_8","alias_value":"35254I5X","created_at":"2026-07-10T01:18:58Z"}],"graph_snapshots":[{"event_id":"sha256:6a9a504b7106f8296944630f1d496df1c0337cb67b90386de5a737486785855f","target":"graph","created_at":"2026-07-10T01:18:58Z","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/2603.12478/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal instruction tuning is often compute-inefficient because training budgets are spread across large mixed image-video pools whose utility is highly uneven. We present Goal-Driven Data Optimization (GDO), a framework that computes six sample descriptors for each candidate and constructs optimized 1$\\times$ training subsets for different goals. Under a fixed one-epoch Qwen3-VL-8B-Instruct training and evaluation recipe on 8 H20 GPUs, GDO uses far fewer training samples than the Uni-10x baseline while converging faster and achieving higher accuracy. Relative to the fixed 512k-sample Uni-1","authors_text":"Hai Ci, Haozhe Zhao, Rujie Wu, Yizhou Wang","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-12T21:54:50Z","title":"Less Data, Faster Convergence: Goal-Driven Data Optimization for Multimodal Instruction Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.12478","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":"sha256:3b469953d28feb1fead1effc1d4db6fa9626f68fa785b52107b1a2bd33b68923","target":"record","created_at":"2026-07-10T01:18:58Z","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":"64f0e75e59b9bc92fba4e1b9c6408393b6ae70d0046af2932782ffdc0f539006","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-12T21:54:50Z","title_canon_sha256":"4ad27a4050331b6b74659e510b5d9e9a78fff4d54b4bfe53787e75100a3acdee"},"schema_version":"1.0","source":{"id":"2603.12478","kind":"arxiv","version":2}},"canonical_sha256":"df75de23b743e37644ebcc748f2d1cf035066eb1335d5694a0f5f1a0662b261c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"df75de23b743e37644ebcc748f2d1cf035066eb1335d5694a0f5f1a0662b261c","first_computed_at":"2026-07-10T01:18:58.867623Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T01:18:58.867623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vCKVaA4rgTDqVLs4GvgNSiMAHx8iIZcxBfkig3sCwn3L70zHfEt+IstLKKxnhNOJiXwNI7NUJafHD5l7v8oUDw==","signature_status":"signed_v1","signed_at":"2026-07-10T01:18:58.868121Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.12478","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3b469953d28feb1fead1effc1d4db6fa9626f68fa785b52107b1a2bd33b68923","sha256:6a9a504b7106f8296944630f1d496df1c0337cb67b90386de5a737486785855f"],"state_sha256":"525aa67647200c772fe25038f5b9bc3b534f1a85f6cc4171c09936b7d2bf6e12"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fkkem7DfjUfQ6mAWKHsqYCBJSuT4XJ04pEfLRR99WO89qRIUsMvrf5OXlrfkgVDGe6MVhR1IRjVkC4JSL9gkDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T12:02:13.315054Z","bundle_sha256":"58096b746e8114f3d3652faa19b9029a4c19f5c3ba546d60e2a8f0429d559227"}}