{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YMJNFF3GKDF5ITPVHOUQKO6D7J","short_pith_number":"pith:YMJNFF3G","schema_version":"1.0","canonical_sha256":"c312d2976650cbd44df53ba9053bc3fa6004c1ba2fd631103acb5c432caaf683","source":{"kind":"arxiv","id":"2604.09482","version":2},"attestation_state":"computed","paper":{"title":"Process Reward Agents for Steering Knowledge-Intensive Reasoning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Process Reward Agents supply online step-wise rewards from external knowledge to steer reasoning in frozen language models.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jiwoong Sohn, Kenneth Styppa, Michael Moor, Tomasz Sternal, Torsten Hoefler","submitted_at":"2026-04-10T16:45:44Z","abstract_excerpt":"Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Rewa"},"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":"2604.09482","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-10T16:45:44Z","cross_cats_sorted":[],"title_canon_sha256":"92b491970cb747dc463d80ee2cb8307cf3a866ffd7de9570096323fae0af1e83","abstract_canon_sha256":"3d68182b83fcba80a276f656f3b39509f5e035d32b9d9de46c8f8c9f974d5d13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:52.913585Z","signature_b64":"Jdou72kvOTC3SEDtheQhEtmt3rzIXNnMZwuBiv+eVszjhKUxm4C4L66v1wGgBCK5hJ+V7qzMDeu1e+lZ56X+Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c312d2976650cbd44df53ba9053bc3fa6004c1ba2fd631103acb5c432caaf683","last_reissued_at":"2026-06-02T02:04:52.913127Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:52.913127Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Process Reward Agents for Steering Knowledge-Intensive Reasoning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Process Reward Agents supply online step-wise rewards from external knowledge to steer reasoning in frozen language models.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jiwoong Sohn, Kenneth Styppa, Michael Moor, Tomasz Sternal, Torsten Hoefler","submitted_at":"2026-04-10T16:45:44Z","abstract_excerpt":"Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Rewa"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"PRA consistently outperforms strong baselines, achieving 80.8% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That retrieval-augmented process rewards can be computed reliably and cheaply at every generation step from external knowledge sources without introducing undetected errors or prohibitive latency that would negate the search benefits.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Process Reward Agents enable online step-wise guidance for frozen AI models in medical reasoning, raising accuracy to 80.8% on MedQA and up to 25.7% gains across 0.5B-8B models without policy updates.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Process Reward Agents supply online step-wise rewards from external knowledge to steer reasoning in frozen language models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bf2bc304333b3d191247118c6c200e1e5447a6369ea4569439463b0c149ac691"},"source":{"id":"2604.09482","kind":"arxiv","version":2},"verdict":{"id":"5aff9e65-e160-4734-9b40-00819bb5c476","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T17:57:44.951995Z","strongest_claim":"PRA consistently outperforms strong baselines, achieving 80.8% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates.","one_line_summary":"Process Reward Agents enable online step-wise guidance for frozen AI models in medical reasoning, raising accuracy to 80.8% on MedQA and up to 25.7% gains across 0.5B-8B models without policy updates.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That retrieval-augmented process rewards can be computed reliably and cheaply at every generation step from external knowledge sources without introducing undetected errors or prohibitive latency that would negate the search benefits.","pith_extraction_headline":"Process Reward Agents supply online step-wise rewards from external knowledge to steer reasoning in frozen language models."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.09482/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":"2604.09482","created_at":"2026-06-02T02:04:52.913185+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.09482v2","created_at":"2026-06-02T02:04:52.913185+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.09482","created_at":"2026-06-02T02:04:52.913185+00:00"},{"alias_kind":"pith_short_12","alias_value":"YMJNFF3GKDF5","created_at":"2026-06-02T02:04:52.913185+00:00"},{"alias_kind":"pith_short_16","alias_value":"YMJNFF3GKDF5ITPV","created_at":"2026-06-02T02:04:52.913185+00:00"},{"alias_kind":"pith_short_8","alias_value":"YMJNFF3G","created_at":"2026-06-02T02:04:52.913185+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/YMJNFF3GKDF5ITPVHOUQKO6D7J","json":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J.json","graph_json":"https://pith.science/api/pith-number/YMJNFF3GKDF5ITPVHOUQKO6D7J/graph.json","events_json":"https://pith.science/api/pith-number/YMJNFF3GKDF5ITPVHOUQKO6D7J/events.json","paper":"https://pith.science/paper/YMJNFF3G"},"agent_actions":{"view_html":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J","download_json":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J.json","view_paper":"https://pith.science/paper/YMJNFF3G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.09482&json=true","fetch_graph":"https://pith.science/api/pith-number/YMJNFF3GKDF5ITPVHOUQKO6D7J/graph.json","fetch_events":"https://pith.science/api/pith-number/YMJNFF3GKDF5ITPVHOUQKO6D7J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J/action/storage_attestation","attest_author":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J/action/author_attestation","sign_citation":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J/action/citation_signature","submit_replication":"https://pith.science/pith/YMJNFF3GKDF5ITPVHOUQKO6D7J/action/replication_record"}},"created_at":"2026-06-02T02:04:52.913185+00:00","updated_at":"2026-06-02T02:04:52.913185+00:00"}