{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZYL2MQ34FJF4PDV5I2F5DP3DI7","short_pith_number":"pith:ZYL2MQ34","schema_version":"1.0","canonical_sha256":"ce17a6437c2a4bc78ebd468bd1bf6347f921231d507229e4e0ede542c69d6c1f","source":{"kind":"arxiv","id":"2606.03239","version":1},"attestation_state":"computed","paper":{"title":"ARBOR: Online Process Rewards via a Reusable Rubric Buffer for Search Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengfu Huo, Liang Ding, Longxiang Zhang, Shaoxiong Zhan, Shu-Tao Xia, Tao Dai, Wen Huang, Xin Shan, Xintong Wang, Zheng Liu, Zhiang Xu","submitted_at":"2026-06-02T06:58:54Z","abstract_excerpt":"LLM-based search agents are trained predominantly with outcome-only reward, leaving the search process itself unsupervised. This signal degenerates on outcome-homogeneous groups where all sampled trajectories share the same correctness, yielding zero within-group advantage and no gradient. Existing process supervision either trains a costly verifier or generates per-query rubrics that are inconsistent across queries and discarded after one use. We propose ARBOR (Adaptive Rubric Buffer for Online Reward), a reusable process-reward framework that maintains a rubric memory shared across queries. "},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.03239","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-02T06:58:54Z","cross_cats_sorted":[],"title_canon_sha256":"caf2102f3b6053142719ff907f2117dfbe97c0517020fa6c9884423bb96eb346","abstract_canon_sha256":"417109bfdc2b2ad2240274ee2a3d892d6303d9c727848744fbb0b3f81fb73e50"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:35.860332Z","signature_b64":"lBQ1KlO6gGVREjv77Px76LxSrPAZtBDtu2RK7yvfR6e6poF6aIJG8eU5BEiQOj1mdvjCCGR1MhuhpkODnIF1Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce17a6437c2a4bc78ebd468bd1bf6347f921231d507229e4e0ede542c69d6c1f","last_reissued_at":"2026-06-03T01:05:35.859905Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:35.859905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ARBOR: Online Process Rewards via a Reusable Rubric Buffer for Search Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengfu Huo, Liang Ding, Longxiang Zhang, Shaoxiong Zhan, Shu-Tao Xia, Tao Dai, Wen Huang, Xin Shan, Xintong Wang, Zheng Liu, Zhiang Xu","submitted_at":"2026-06-02T06:58:54Z","abstract_excerpt":"LLM-based search agents are trained predominantly with outcome-only reward, leaving the search process itself unsupervised. This signal degenerates on outcome-homogeneous groups where all sampled trajectories share the same correctness, yielding zero within-group advantage and no gradient. Existing process supervision either trains a costly verifier or generates per-query rubrics that are inconsistent across queries and discarded after one use. We propose ARBOR (Adaptive Rubric Buffer for Online Reward), a reusable process-reward framework that maintains a rubric memory shared across queries. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03239","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/2606.03239/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.03239","created_at":"2026-06-03T01:05:35.859967+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03239v1","created_at":"2026-06-03T01:05:35.859967+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03239","created_at":"2026-06-03T01:05:35.859967+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYL2MQ34FJF4","created_at":"2026-06-03T01:05:35.859967+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYL2MQ34FJF4PDV5","created_at":"2026-06-03T01:05:35.859967+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYL2MQ34","created_at":"2026-06-03T01:05:35.859967+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7","json":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7.json","graph_json":"https://pith.science/api/pith-number/ZYL2MQ34FJF4PDV5I2F5DP3DI7/graph.json","events_json":"https://pith.science/api/pith-number/ZYL2MQ34FJF4PDV5I2F5DP3DI7/events.json","paper":"https://pith.science/paper/ZYL2MQ34"},"agent_actions":{"view_html":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7","download_json":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7.json","view_paper":"https://pith.science/paper/ZYL2MQ34","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03239&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYL2MQ34FJF4PDV5I2F5DP3DI7/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYL2MQ34FJF4PDV5I2F5DP3DI7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7/action/storage_attestation","attest_author":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7/action/author_attestation","sign_citation":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7/action/citation_signature","submit_replication":"https://pith.science/pith/ZYL2MQ34FJF4PDV5I2F5DP3DI7/action/replication_record"}},"created_at":"2026-06-03T01:05:35.859967+00:00","updated_at":"2026-06-03T01:05:35.859967+00:00"}