{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:R5NDLVTF6LLJQTXA4BVGEQO3MP","short_pith_number":"pith:R5NDLVTF","canonical_record":{"source":{"id":"2504.10559","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-14T14:53:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7afb15eb9aee1769a13cf26db94e9ecee93530e6aabcfbc072765195af9f24db","abstract_canon_sha256":"b0e5520408a4b9814e33200039d6b612efeb7e00a2281fdf393241cbcbd0388e"},"schema_version":"1.0"},"canonical_sha256":"8f5a35d665f2d6984ee0e06a6241db63c3ebc203b1248f978d7e1b7adf7c2fac","source":{"kind":"arxiv","id":"2504.10559","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.10559","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"arxiv_version","alias_value":"2504.10559v1","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10559","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_12","alias_value":"R5NDLVTF6LLJ","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_16","alias_value":"R5NDLVTF6LLJQTXA","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_8","alias_value":"R5NDLVTF","created_at":"2026-07-05T10:49:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:R5NDLVTF6LLJQTXA4BVGEQO3MP","target":"record","payload":{"canonical_record":{"source":{"id":"2504.10559","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-14T14:53:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7afb15eb9aee1769a13cf26db94e9ecee93530e6aabcfbc072765195af9f24db","abstract_canon_sha256":"b0e5520408a4b9814e33200039d6b612efeb7e00a2281fdf393241cbcbd0388e"},"schema_version":"1.0"},"canonical_sha256":"8f5a35d665f2d6984ee0e06a6241db63c3ebc203b1248f978d7e1b7adf7c2fac","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:19.669506Z","signature_b64":"NLzkTSvrAChlsiIDqmepOQTcxzRA/JoHj+gk8qsp7ZzNm0SYBgL74v8jw5z1+JIZvlUB2fW9giqfft46i/+ACg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f5a35d665f2d6984ee0e06a6241db63c3ebc203b1248f978d7e1b7adf7c2fac","last_reissued_at":"2026-07-05T10:49:19.668983Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:19.668983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.10559","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":"2026-07-05T10:49:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mzpr8FkxIhynqDmGfSTNIxs5uLe8Yg44gjsSHqtuf18IsoXZ7hryBXCEfXjFKYEf9sjD3d6Jj81TPibboa+0Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:28:26.339585Z"},"content_sha256":"6229533b8f838b68c03d07a6632774ac252f6eb07816ac9d87305d4150a953ff","schema_version":"1.0","event_id":"sha256:6229533b8f838b68c03d07a6632774ac252f6eb07816ac9d87305d4150a953ff"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:R5NDLVTF6LLJQTXA4BVGEQO3MP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Process Reward Model Training via Active Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Changyu Chen, Keyu Duan, Longxu Dou, Michael Qizhe Shieh, Qiguang Chen, Tianyu Pang, Xin Mao, Zichen Liu","submitted_at":"2025-04-14T14:53:56Z","abstract_excerpt":"Process Reward Models (PRMs) provide step-level supervision to large language models (LLMs), but scaling up training data annotation remains challenging for both humans and LLMs. To address this limitation, we propose an active learning approach, ActPRM, which proactively selects the most uncertain samples for training, substantially reducing labeling costs. During training, we use the PRM to estimate uncertainty after the forward pass, retaining only highly uncertain data. A capable yet costly reasoning model then labels this data. Then we compute the loss with respect to the labels and updat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10559","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/2504.10559/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":"2026-07-05T10:49:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RVRY0Fx9PlqsjhEppmQs3PeqjU0AukwyGihiohEf2DvmDhvtioZ6s0djYdNtCAHIlV5am0GIHNNKZ4MvmU67Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:28:26.339957Z"},"content_sha256":"8ba7b2ef7bce52c36916bae86c1c6c671691552071d03ea2dca2239ea8d5b2be","schema_version":"1.0","event_id":"sha256:8ba7b2ef7bce52c36916bae86c1c6c671691552071d03ea2dca2239ea8d5b2be"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/bundle.json","state_url":"https://pith.science/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/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":"2026-07-07T12:28:26Z","links":{"resolver":"https://pith.science/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP","bundle":"https://pith.science/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/bundle.json","state":"https://pith.science/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R5NDLVTF6LLJQTXA4BVGEQO3MP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:R5NDLVTF6LLJQTXA4BVGEQO3MP","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":"b0e5520408a4b9814e33200039d6b612efeb7e00a2281fdf393241cbcbd0388e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-14T14:53:56Z","title_canon_sha256":"7afb15eb9aee1769a13cf26db94e9ecee93530e6aabcfbc072765195af9f24db"},"schema_version":"1.0","source":{"id":"2504.10559","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.10559","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"arxiv_version","alias_value":"2504.10559v1","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10559","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_12","alias_value":"R5NDLVTF6LLJ","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_16","alias_value":"R5NDLVTF6LLJQTXA","created_at":"2026-07-05T10:49:19Z"},{"alias_kind":"pith_short_8","alias_value":"R5NDLVTF","created_at":"2026-07-05T10:49:19Z"}],"graph_snapshots":[{"event_id":"sha256:8ba7b2ef7bce52c36916bae86c1c6c671691552071d03ea2dca2239ea8d5b2be","target":"graph","created_at":"2026-07-05T10:49:19Z","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/2504.10559/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Process Reward Models (PRMs) provide step-level supervision to large language models (LLMs), but scaling up training data annotation remains challenging for both humans and LLMs. To address this limitation, we propose an active learning approach, ActPRM, which proactively selects the most uncertain samples for training, substantially reducing labeling costs. During training, we use the PRM to estimate uncertainty after the forward pass, retaining only highly uncertain data. A capable yet costly reasoning model then labels this data. Then we compute the loss with respect to the labels and updat","authors_text":"Changyu Chen, Keyu Duan, Longxu Dou, Michael Qizhe Shieh, Qiguang Chen, Tianyu Pang, Xin Mao, Zichen Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-14T14:53:56Z","title":"Efficient Process Reward Model Training via Active Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10559","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":"sha256:6229533b8f838b68c03d07a6632774ac252f6eb07816ac9d87305d4150a953ff","target":"record","created_at":"2026-07-05T10:49:19Z","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":"b0e5520408a4b9814e33200039d6b612efeb7e00a2281fdf393241cbcbd0388e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-14T14:53:56Z","title_canon_sha256":"7afb15eb9aee1769a13cf26db94e9ecee93530e6aabcfbc072765195af9f24db"},"schema_version":"1.0","source":{"id":"2504.10559","kind":"arxiv","version":1}},"canonical_sha256":"8f5a35d665f2d6984ee0e06a6241db63c3ebc203b1248f978d7e1b7adf7c2fac","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8f5a35d665f2d6984ee0e06a6241db63c3ebc203b1248f978d7e1b7adf7c2fac","first_computed_at":"2026-07-05T10:49:19.668983Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:49:19.668983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NLzkTSvrAChlsiIDqmepOQTcxzRA/JoHj+gk8qsp7ZzNm0SYBgL74v8jw5z1+JIZvlUB2fW9giqfft46i/+ACg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:49:19.669506Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.10559","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6229533b8f838b68c03d07a6632774ac252f6eb07816ac9d87305d4150a953ff","sha256:8ba7b2ef7bce52c36916bae86c1c6c671691552071d03ea2dca2239ea8d5b2be"],"state_sha256":"0aa4aab0b706995a5b31109e9ac07c6c5335846431034f18657551dacedb80a3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EbRZxL2UvcKOOL7yQ/jSgBdshEG5huhnqOgmpsg6PumhSlRi6vo66afITxxxSvAEjt71653wqR7m1lPQc909Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:28:26.341812Z","bundle_sha256":"b0c78b6bab9d4437ae48f874a5e7af37ea1e0e3caf442b1cfc4b2ff1448b0e8e"}}