{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OPXDWCNM7X4UVFEFCZQM4PDPKM","short_pith_number":"pith:OPXDWCNM","canonical_record":{"source":{"id":"2605.14373","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T04:52:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c3e3bab63ffae11e3e791827d4b003b1e730f4b04601a03d236eb04373780a25","abstract_canon_sha256":"9d4e3544c50b79d1b43ceae5aa1b66a308b0e3adb0da816190cc102fef8d78bf"},"schema_version":"1.0"},"canonical_sha256":"73ee3b09acfdf94a94851660ce3c6f531acbd98c406776752a89bb808c6d6732","source":{"kind":"arxiv","id":"2605.14373","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14373","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14373v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14373","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"OPXDWCNM7X4U","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"OPXDWCNM7X4UVFEF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"OPXDWCNM","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OPXDWCNM7X4UVFEFCZQM4PDPKM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14373","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T04:52:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c3e3bab63ffae11e3e791827d4b003b1e730f4b04601a03d236eb04373780a25","abstract_canon_sha256":"9d4e3544c50b79d1b43ceae5aa1b66a308b0e3adb0da816190cc102fef8d78bf"},"schema_version":"1.0"},"canonical_sha256":"73ee3b09acfdf94a94851660ce3c6f531acbd98c406776752a89bb808c6d6732","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:07.805133Z","signature_b64":"vN3TABzyntmeNK40/wp29gzhqoLQr8E7TOJ9Q2D/bEHf9hFYY/Q3kCVOA1m1noF03zKIzSEY7oAgXAUo9NIyCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"73ee3b09acfdf94a94851660ce3c6f531acbd98c406776752a89bb808c6d6732","last_reissued_at":"2026-05-17T23:39:07.804599Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:07.804599Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14373","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":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4yWbQQUB0srpQ4SJORWMVkw7AAFGJX/tjM1Z8bIOf4dRKJIZcKGMfcbsvo7zyz2C0dv2tppaq88VngC+vhB6Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T12:21:38.485217Z"},"content_sha256":"d110336e7ea1df38dcdab71595bdfa442228d1aa46372d42c15792b453c2d662","schema_version":"1.0","event_id":"sha256:d110336e7ea1df38dcdab71595bdfa442228d1aa46372d42c15792b453c2d662"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OPXDWCNM7X4UVFEFCZQM4PDPKM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chen Liang, Daniel Rakita, Qian Wang, Xiatao Sun","submitted_at":"2026-05-14T04:52:24Z","abstract_excerpt":"Zeroth-Order (ZO) optimization is pivotal for scenarios where backpropagation is unavailable, such as memory-constrained on-device learning and black-box optimization. However, existing methods face a stark trade-off: they are either sample-inefficient (e.g., standard finite differences) or suffer from high variance due to randomized estimation (e.g., random subspace methods). In this work, we propose Coherent Coordinate Descent (CoCD), a deterministic, sample-efficient, and budget-aware ZO optimizer. Theoretically, we formalize the notion of gradient coherence and demonstrate that CoCD is equ"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CoCD is equivalent to Block Cyclic Coordinate Descent with warm starts, enabling O(1) query complexity per step while maintaining global descent directions; larger finite-difference step sizes induce implicit smoothing by reducing the effective smoothness constant.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Historical gradients remain sufficiently coherent across iterations to provide reliable descent directions without significant landscape drift or loss of global information.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CoCD converts stale gradients into stable descent directions for zeroth-order optimization through coherent coordinate updates and implicit landscape smoothing from larger finite-difference steps.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d8694eb94551951c57f618bdc744adf00773aa92ae4dc790f52fe0f75aed57af"},"source":{"id":"2605.14373","kind":"arxiv","version":1},"verdict":{"id":"b49b0968-e78e-471a-a011-47d93936aa19","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:44:56.481353Z","strongest_claim":"CoCD is equivalent to Block Cyclic Coordinate Descent with warm starts, enabling O(1) query complexity per step while maintaining global descent directions; larger finite-difference step sizes induce implicit smoothing by reducing the effective smoothness constant.","one_line_summary":"CoCD converts stale gradients into stable descent directions for zeroth-order optimization through coherent coordinate updates and implicit landscape smoothing from larger finite-difference steps.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Historical gradients remain sufficiently coherent across iterations to provide reliable descent directions without significant landscape drift or loss of global information.","pith_extraction_headline":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization."},"references":{"count":35,"sample":[{"doi":"","year":null,"title":"arXiv preprint arXiv:2504.18790 , year=","work_id":"3c7632fc-90d6-4779-9216-c2b09698a372","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"IEEE Signal Processing Magazine , volume=","work_id":"762edb74-3ffb-42f2-b862-20ce9c7b53c6","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"International Conference on Artificial Intelligence and Statistics , pages=","work_id":"e768ec45-ba21-4b9d-9138-244159f02861","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"International Conference on Learning Representations , year=","work_id":"21014bdc-df06-4015-bfc6-a07ae0632bef","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"International Conference on Machine Learning , pages=","work_id":"e702eb44-94af-4897-b2fe-71e567898432","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":35,"snapshot_sha256":"d5b0ab5462bfa0c62a22d8bddd40d1338a565af7fd510bd6d408f54a87764cc9","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"3118bb8ec2aee895e0a30d0023add0e8459c1e9758ea85d8885de6d47a165d08"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"b49b0968-e78e-471a-a011-47d93936aa19"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+vy6TZEQKUwMfDtD084chih6qhEbB58fgjqsifHIOdllEsUczIHiRZkAl0XLtC2v9OOB4c/6pZ5rHAVIkUUeDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T12:21:38.486170Z"},"content_sha256":"e3dc2103d79fe8fd599e95a7a3926c68285db0b66385ecb3ba987144ba61de15","schema_version":"1.0","event_id":"sha256:e3dc2103d79fe8fd599e95a7a3926c68285db0b66385ecb3ba987144ba61de15"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/bundle.json","state_url":"https://pith.science/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/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":"2026-06-06T12:21:38Z","links":{"resolver":"https://pith.science/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM","bundle":"https://pith.science/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/bundle.json","state":"https://pith.science/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OPXDWCNM7X4UVFEFCZQM4PDPKM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OPXDWCNM7X4UVFEFCZQM4PDPKM","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":"9d4e3544c50b79d1b43ceae5aa1b66a308b0e3adb0da816190cc102fef8d78bf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T04:52:24Z","title_canon_sha256":"c3e3bab63ffae11e3e791827d4b003b1e730f4b04601a03d236eb04373780a25"},"schema_version":"1.0","source":{"id":"2605.14373","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14373","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14373v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14373","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"OPXDWCNM7X4U","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"OPXDWCNM7X4UVFEF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"OPXDWCNM","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:e3dc2103d79fe8fd599e95a7a3926c68285db0b66385ecb3ba987144ba61de15","target":"graph","created_at":"2026-05-17T23:39:07Z","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":"CoCD is equivalent to Block Cyclic Coordinate Descent with warm starts, enabling O(1) query complexity per step while maintaining global descent directions; larger finite-difference step sizes induce implicit smoothing by reducing the effective smoothness constant."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"Historical gradients remain sufficiently coherent across iterations to provide reliable descent directions without significant landscape drift or loss of global information."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"CoCD converts stale gradients into stable descent directions for zeroth-order optimization through coherent coordinate updates and implicit landscape smoothing from larger finite-difference steps."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization."}],"snapshot_sha256":"d8694eb94551951c57f618bdc744adf00773aa92ae4dc790f52fe0f75aed57af"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"3118bb8ec2aee895e0a30d0023add0e8459c1e9758ea85d8885de6d47a165d08"},"paper":{"abstract_excerpt":"Zeroth-Order (ZO) optimization is pivotal for scenarios where backpropagation is unavailable, such as memory-constrained on-device learning and black-box optimization. However, existing methods face a stark trade-off: they are either sample-inefficient (e.g., standard finite differences) or suffer from high variance due to randomized estimation (e.g., random subspace methods). In this work, we propose Coherent Coordinate Descent (CoCD), a deterministic, sample-efficient, and budget-aware ZO optimizer. Theoretically, we formalize the notion of gradient coherence and demonstrate that CoCD is equ","authors_text":"Chen Liang, Daniel Rakita, Qian Wang, Xiatao Sun","cross_cats":["cs.AI"],"headline":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T04:52:24Z","title":"Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization"},"references":{"count":35,"internal_anchors":1,"resolved_work":35,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"arXiv preprint arXiv:2504.18790 , year=","work_id":"3c7632fc-90d6-4779-9216-c2b09698a372","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"IEEE Signal Processing Magazine , volume=","work_id":"762edb74-3ffb-42f2-b862-20ce9c7b53c6","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"International Conference on Artificial Intelligence and Statistics , pages=","work_id":"e768ec45-ba21-4b9d-9138-244159f02861","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"International Conference on Learning Representations , year=","work_id":"21014bdc-df06-4015-bfc6-a07ae0632bef","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"International Conference on Machine Learning , pages=","work_id":"e702eb44-94af-4897-b2fe-71e567898432","year":2018}],"snapshot_sha256":"d5b0ab5462bfa0c62a22d8bddd40d1338a565af7fd510bd6d408f54a87764cc9"},"source":{"id":"2605.14373","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:44:56.481353Z","id":"b49b0968-e78e-471a-a011-47d93936aa19","model_set":{"reader":"grok-4.3"},"one_line_summary":"CoCD converts stale gradients into stable descent directions for zeroth-order optimization through coherent coordinate updates and implicit landscape smoothing from larger finite-difference steps.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Coherent Coordinate Descent reuses historical gradients as warm starts to achieve O(1) query cost while keeping global descent directions in zeroth-order optimization.","strongest_claim":"CoCD is equivalent to Block Cyclic Coordinate Descent with warm starts, enabling O(1) query complexity per step while maintaining global descent directions; larger finite-difference step sizes induce implicit smoothing by reducing the effective smoothness constant.","weakest_assumption":"Historical gradients remain sufficiently coherent across iterations to provide reliable descent directions without significant landscape drift or loss of global information."}},"verdict_id":"b49b0968-e78e-471a-a011-47d93936aa19"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d110336e7ea1df38dcdab71595bdfa442228d1aa46372d42c15792b453c2d662","target":"record","created_at":"2026-05-17T23:39:07Z","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":"9d4e3544c50b79d1b43ceae5aa1b66a308b0e3adb0da816190cc102fef8d78bf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T04:52:24Z","title_canon_sha256":"c3e3bab63ffae11e3e791827d4b003b1e730f4b04601a03d236eb04373780a25"},"schema_version":"1.0","source":{"id":"2605.14373","kind":"arxiv","version":1}},"canonical_sha256":"73ee3b09acfdf94a94851660ce3c6f531acbd98c406776752a89bb808c6d6732","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"73ee3b09acfdf94a94851660ce3c6f531acbd98c406776752a89bb808c6d6732","first_computed_at":"2026-05-17T23:39:07.804599Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:07.804599Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vN3TABzyntmeNK40/wp29gzhqoLQr8E7TOJ9Q2D/bEHf9hFYY/Q3kCVOA1m1noF03zKIzSEY7oAgXAUo9NIyCQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:07.805133Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14373","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d110336e7ea1df38dcdab71595bdfa442228d1aa46372d42c15792b453c2d662","sha256:e3dc2103d79fe8fd599e95a7a3926c68285db0b66385ecb3ba987144ba61de15"],"state_sha256":"bb5ee2d3be83a1522a320875dd4727cb64201a1096381e336c6b40a9b90a07af"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iUejgXqnBmSh6inrAZrDuobCPqDuII8RReEnSTD5WY3uFRStO6fGa04at3Urhn120GgfbRHhKFYfARBrej+MCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T12:21:38.490489Z","bundle_sha256":"9d930fe7095017267ce6449c201dc95b36a483e201d57a08a6dd6f960276f858"}}