{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DRRUJNA2MNQ5FTSSRA74WE5PVW","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":"ad3fd56b0337e4767d78b763ec3dc90a9f4b8d0ca658ae3e3d0b8a5f7124f420","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T03:55:12Z","title_canon_sha256":"e028ad2b04f7a6bf7c642840e73e0505b7bf462617a717256e69d84b17bbc50d"},"schema_version":"1.0","source":{"id":"2605.11480","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11480","created_at":"2026-05-20T00:03:17Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11480v2","created_at":"2026-05-20T00:03:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11480","created_at":"2026-05-20T00:03:17Z"},{"alias_kind":"pith_short_12","alias_value":"DRRUJNA2MNQ5","created_at":"2026-05-20T00:03:17Z"},{"alias_kind":"pith_short_16","alias_value":"DRRUJNA2MNQ5FTSS","created_at":"2026-05-20T00:03:17Z"},{"alias_kind":"pith_short_8","alias_value":"DRRUJNA2","created_at":"2026-05-20T00:03:17Z"}],"graph_snapshots":[{"event_id":"sha256:0440e5b048a5700e071bfcb778492770d23baa042e4aecfcfdf5ae79d03a7acd","target":"graph","created_at":"2026-05-20T00:03:17Z","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":"On standard text-to-image reward fine-tuning benchmarks, EAM converges up to 4x faster than AM and matches or surpasses it across various metrics including PickScore, ImageReward, HPSv2.1, CLIPScore and Aesthetics."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That reformulating the SOC problem with linear base drift and modified terminal cost preserves the original alignment objective and solution quality without introducing bias or loss of expressivity."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"EAM speeds up adjoint matching for diffusion model reward fine-tuning by switching to linear base drift, allowing deterministic few-step solvers and closed-form adjoints with up to 4x faster convergence on text-to-image benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"By reformulating the stochastic optimal control problem with linear base drift, Efficient Adjoint Matching speeds up reward fine-tuning of diffusion models up to 4x while matching performance."}],"snapshot_sha256":"961c7d085c483c52d90646cb223e319207fdb86ce2ae1cd1ed15150b6693adca"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8405a874ae7dfca82d3f39c9630be4e76245a017cfbd2d5282edded404646bb0"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T12:35:13.612761Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T09:31:19.317994Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:21:51.152392Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11480/integrity.json","findings":[],"snapshot_sha256":"d69ab342a34e23f0fa3d59dbf3f9abaaeb4205df9b42b33269cb10e1ad50b6d3","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward fine-tuning has become a common approach for aligning pretrained diffusion and flow models with human preferences in text-to-image generation. Among reward-gradient-based methods, Adjoint Matching (AM) provides a principled formulation by casting reward fine-tuning as a stochastic optimal control (SOC) problem. However, AM inevitably requires a substantial computational cost: it requires (i) stochastic simulation of full generative trajectories under memoryless dynamics, resulting in a large number of function evaluations, and (ii) backward ODE simulation of the adjoint state along each","authors_text":"Dongsoo Shin, Jaemoo Choi, Jaewoong Choi, Jeongwoo Shin, Joonseok Lee, Wei Guo, Yongxin Chen, Yuchen Zhu","cross_cats":[],"headline":"By reformulating the stochastic optimal control problem with linear base drift, Efficient Adjoint Matching speeds up reward fine-tuning of diffusion models up to 4x while matching performance.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T03:55:12Z","title":"Efficient Adjoint Matching for Fine-tuning Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11480","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T01:48:08.979614Z","id":"c0d06026-e430-4267-9e7d-ca052735b6cd","model_set":{"reader":"grok-4.3"},"one_line_summary":"EAM speeds up adjoint matching for diffusion model reward fine-tuning by switching to linear base drift, allowing deterministic few-step solvers and closed-form adjoints with up to 4x faster convergence on text-to-image benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"By reformulating the stochastic optimal control problem with linear base drift, Efficient Adjoint Matching speeds up reward fine-tuning of diffusion models up to 4x while matching performance.","strongest_claim":"On standard text-to-image reward fine-tuning benchmarks, EAM converges up to 4x faster than AM and matches or surpasses it across various metrics including PickScore, ImageReward, HPSv2.1, CLIPScore and Aesthetics.","weakest_assumption":"That reformulating the SOC problem with linear base drift and modified terminal cost preserves the original alignment objective and solution quality without introducing bias or loss of expressivity."}},"verdict_id":"c0d06026-e430-4267-9e7d-ca052735b6cd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a791a7d296c258c7a9dfb42dbca0f49f00797ee103945ac8e9028e4d7314d8eb","target":"record","created_at":"2026-05-20T00:03:17Z","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":"ad3fd56b0337e4767d78b763ec3dc90a9f4b8d0ca658ae3e3d0b8a5f7124f420","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T03:55:12Z","title_canon_sha256":"e028ad2b04f7a6bf7c642840e73e0505b7bf462617a717256e69d84b17bbc50d"},"schema_version":"1.0","source":{"id":"2605.11480","kind":"arxiv","version":2}},"canonical_sha256":"1c6344b41a6361d2ce52883fcb13afada8e98f83b41346cb3f23dfa2d25d79cf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1c6344b41a6361d2ce52883fcb13afada8e98f83b41346cb3f23dfa2d25d79cf","first_computed_at":"2026-05-20T00:03:17.800309Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:17.800309Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kzIHdJHIhD2Eo5ApozkV+0L73WjsyeK39ONkA7y3rKQdyB8yKd1qRy9OUmy1GgBP7Iuu6NG32q7dtwY9V4R/CA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:17.801137Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11480","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a791a7d296c258c7a9dfb42dbca0f49f00797ee103945ac8e9028e4d7314d8eb","sha256:0440e5b048a5700e071bfcb778492770d23baa042e4aecfcfdf5ae79d03a7acd"],"state_sha256":"f9fac6159067ece022af28614e2224a705ec377a532bdae0a2f3669e40ab1704"}