{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Q4F7RUBKK2IZHFPIAJTVEUKXGM","short_pith_number":"pith:Q4F7RUBK","schema_version":"1.0","canonical_sha256":"870bf8d02a56919395e802675251573321a41d43e4aff08d897bd86a8641ff65","source":{"kind":"arxiv","id":"2605.13130","version":1},"attestation_state":"computed","paper":{"title":"GRACE: Gradient-aligned Reasoning Data Curation for Efficient Post-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"GRACE scores each reasoning step by its alignment with the answer gradient and trajectory consistency to select data subsets that match or exceed full performance with 5-20 percent of the samples.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jianghong Ma, Junjie Li, Ningxuan Ma, Xiaofeng Zhang, Ziao Wang","submitted_at":"2026-05-13T07:55:39Z","abstract_excerpt":"Existing reasoning data curation pipelines score whole samples, treating every intermediate step as equally valuable. In reality, steps within a trace contribute very unevenly, and selecting reasoning data well requires assessing them individually. We present GRACE, a gradient-aligned curation method that views each reasoning trace as a sequence of optimization events and scores every step by two complementary signals: its alignment with the answer-oriented gradient direction, and its consistency with the preceding reasoning trajectory. Step-level scores are aggregated into a sample-level valu"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.13130","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T07:55:39Z","cross_cats_sorted":[],"title_canon_sha256":"8d985cead699e72f8e855647819a7df04204444661269631e7f85e2af3ba6cf1","abstract_canon_sha256":"4227439fb8a5dcf0e7f10a0e15349f543aa9bcf7c65d3f3e5d9a89473a50b7bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:08:57.801106Z","signature_b64":"flJfFNGxzYOJqwlqkIqT3Wp+TbmLB9rgTtB9aWCRZXg293k4na14NcbOH2d1ek8uXQF9HR3h5Z76gS4PejZhBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"870bf8d02a56919395e802675251573321a41d43e4aff08d897bd86a8641ff65","last_reissued_at":"2026-05-18T03:08:57.800483Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:08:57.800483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GRACE: Gradient-aligned Reasoning Data Curation for Efficient Post-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"GRACE scores each reasoning step by its alignment with the answer gradient and trajectory consistency to select data subsets that match or exceed full performance with 5-20 percent of the samples.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jianghong Ma, Junjie Li, Ningxuan Ma, Xiaofeng Zhang, Ziao Wang","submitted_at":"2026-05-13T07:55:39Z","abstract_excerpt":"Existing reasoning data curation pipelines score whole samples, treating every intermediate step as equally valuable. In reality, steps within a trace contribute very unevenly, and selecting reasoning data well requires assessing them individually. We present GRACE, a gradient-aligned curation method that views each reasoning trace as a sequence of optimization events and scores every step by two complementary signals: its alignment with the answer-oriented gradient direction, and its consistency with the preceding reasoning trajectory. Step-level scores are aggregated into a sample-level valu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Post-training Qwen3-VL-2B-Instruct on MMathCoT-1M, GRACE reaches 108.8% of the full-data performance with 20% of the data and retains 100.2% with only 5%, with subsets that transfer effectively across model backbones.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the representation-level gradient proxy accurately captures step-level alignment with the answer-oriented gradient and that the two signals (alignment and consistency) reliably identify valuable reasoning steps without external reward models or step annotations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GRACE scores reasoning steps via gradient alignment and trajectory consistency to select data subsets that match full performance with 5% of the data on Qwen3-VL-2B-Instruct.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"GRACE scores each reasoning step by its alignment with the answer gradient and trajectory consistency to select data subsets that match or exceed full performance with 5-20 percent of the samples.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ffc8bbbaaaf523aafc28b6ebae70aa304b391d0a969e5b3b832073fae678949a"},"source":{"id":"2605.13130","kind":"arxiv","version":1},"verdict":{"id":"504e6685-81d7-4c5f-83b5-34d3cfa40f77","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:30:21.939493Z","strongest_claim":"Post-training Qwen3-VL-2B-Instruct on MMathCoT-1M, GRACE reaches 108.8% of the full-data performance with 20% of the data and retains 100.2% with only 5%, with subsets that transfer effectively across model backbones.","one_line_summary":"GRACE scores reasoning steps via gradient alignment and trajectory consistency to select data subsets that match full performance with 5% of the data on Qwen3-VL-2B-Instruct.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the representation-level gradient proxy accurately captures step-level alignment with the answer-oriented gradient and that the two signals (alignment and consistency) reliably identify valuable reasoning steps without external reward models or step annotations.","pith_extraction_headline":"GRACE scores each reasoning step by its alignment with the answer gradient and trajectory consistency to select data subsets that match or exceed full performance with 5-20 percent of the samples."},"references":{"count":37,"sample":[{"doi":"","year":2022,"title":"Chain-of-thought prompting elicits reasoning in large language models","work_id":"3c1378f6-2da6-4177-a291-457b1d9a8feb","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Unlocking mul- timodal mathematical reasoning via process reward model","work_id":"60e34229-23f0-4f66-baf4-54f53416ec28","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"LIMA: less is more for alignment","work_id":"b86e15f2-318c-4e3d-bf76-7d2e60e38819","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V . Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. Self-consistency improves chain of thought reasoning in language models. InThe Elev","work_id":"4c6bab41-df70-4184-9335-14ec662bf3bd","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Mille","work_id":"69ae1ccd-58f3-4ccc-875e-b88f2e78e809","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"515f22013a59ae50b0be7a1c3c4e4f9b7a7d5587198bca1c0c8aad8770aa8900","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"bb03cad62e01311a18ce1ffdf7520798f0b60b56b075330ecb761b567e5a37d9"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.13130","created_at":"2026-05-18T03:08:57.800557+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13130v1","created_at":"2026-05-18T03:08:57.800557+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13130","created_at":"2026-05-18T03:08:57.800557+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q4F7RUBKK2IZ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q4F7RUBKK2IZHFPI","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q4F7RUBK","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM","json":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM.json","graph_json":"https://pith.science/api/pith-number/Q4F7RUBKK2IZHFPIAJTVEUKXGM/graph.json","events_json":"https://pith.science/api/pith-number/Q4F7RUBKK2IZHFPIAJTVEUKXGM/events.json","paper":"https://pith.science/paper/Q4F7RUBK"},"agent_actions":{"view_html":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM","download_json":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM.json","view_paper":"https://pith.science/paper/Q4F7RUBK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13130&json=true","fetch_graph":"https://pith.science/api/pith-number/Q4F7RUBKK2IZHFPIAJTVEUKXGM/graph.json","fetch_events":"https://pith.science/api/pith-number/Q4F7RUBKK2IZHFPIAJTVEUKXGM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM/action/storage_attestation","attest_author":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM/action/author_attestation","sign_citation":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM/action/citation_signature","submit_replication":"https://pith.science/pith/Q4F7RUBKK2IZHFPIAJTVEUKXGM/action/replication_record"}},"created_at":"2026-05-18T03:08:57.800557+00:00","updated_at":"2026-05-18T03:08:57.800557+00:00"}