{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:U5BGOZWALWEG72AAW672F5XTMA","short_pith_number":"pith:U5BGOZWA","schema_version":"1.0","canonical_sha256":"a7426766c05d886fe800b7bfa2f6f3601dd5f3a58ff1ad08f5e2fdea1c9eff20","source":{"kind":"arxiv","id":"2606.26080","version":1},"attestation_state":"computed","paper":{"title":"Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Changdae Oh, Samuel Yeh, Seongheon Park, Sharon Li, Tanwi Mallick, Wendi Li","submitted_at":"2026-06-24T17:54:08Z","abstract_excerpt":"Process reward models enable fine-grained, step-level evaluation of LLMs, yet building them for agentic settings remains prohibitively difficult: long-horizon interactions, irreversible actions, and stochastic environment feedback make both human annotation and Monte Carlo estimation infeasible at scale. In this work, we show that reinforcement learning (RL) post-training already provides the ingredients for effective step-level scoring, eliminating the need for dedicated reward model training altogether. Concretely, we derive an implicit advantage under a general stochastic Markov decision pr"},"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":"2606.26080","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-24T17:54:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73bab3b48aae39d4b7bd05b36c27b0358b2484f3ed7a23ad9ba1df076203c2c5","abstract_canon_sha256":"b1930039963c7305819850320afa3617b73e86047e3ccfbb0e8f972c68e31703"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:48.013530Z","signature_b64":"PpD7WcNmg3EaHbw64njDK3qH9Pmnhzz4JJ4pcn7S7h6c6xFbzkvLKTVJFxNZBK1Xd/1FkM+6Z0iafjAxAWGAAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7426766c05d886fe800b7bfa2f6f3601dd5f3a58ff1ad08f5e2fdea1c9eff20","last_reissued_at":"2026-06-25T01:18:48.013122Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:48.013122Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Changdae Oh, Samuel Yeh, Seongheon Park, Sharon Li, Tanwi Mallick, Wendi Li","submitted_at":"2026-06-24T17:54:08Z","abstract_excerpt":"Process reward models enable fine-grained, step-level evaluation of LLMs, yet building them for agentic settings remains prohibitively difficult: long-horizon interactions, irreversible actions, and stochastic environment feedback make both human annotation and Monte Carlo estimation infeasible at scale. In this work, we show that reinforcement learning (RL) post-training already provides the ingredients for effective step-level scoring, eliminating the need for dedicated reward model training altogether. Concretely, we derive an implicit advantage under a general stochastic Markov decision pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26080","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/2606.26080/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":"2606.26080","created_at":"2026-06-25T01:18:48.013185+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26080v1","created_at":"2026-06-25T01:18:48.013185+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26080","created_at":"2026-06-25T01:18:48.013185+00:00"},{"alias_kind":"pith_short_12","alias_value":"U5BGOZWALWEG","created_at":"2026-06-25T01:18:48.013185+00:00"},{"alias_kind":"pith_short_16","alias_value":"U5BGOZWALWEG72AA","created_at":"2026-06-25T01:18:48.013185+00:00"},{"alias_kind":"pith_short_8","alias_value":"U5BGOZWA","created_at":"2026-06-25T01:18:48.013185+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/U5BGOZWALWEG72AAW672F5XTMA","json":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA.json","graph_json":"https://pith.science/api/pith-number/U5BGOZWALWEG72AAW672F5XTMA/graph.json","events_json":"https://pith.science/api/pith-number/U5BGOZWALWEG72AAW672F5XTMA/events.json","paper":"https://pith.science/paper/U5BGOZWA"},"agent_actions":{"view_html":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA","download_json":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA.json","view_paper":"https://pith.science/paper/U5BGOZWA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26080&json=true","fetch_graph":"https://pith.science/api/pith-number/U5BGOZWALWEG72AAW672F5XTMA/graph.json","fetch_events":"https://pith.science/api/pith-number/U5BGOZWALWEG72AAW672F5XTMA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA/action/storage_attestation","attest_author":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA/action/author_attestation","sign_citation":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA/action/citation_signature","submit_replication":"https://pith.science/pith/U5BGOZWALWEG72AAW672F5XTMA/action/replication_record"}},"created_at":"2026-06-25T01:18:48.013185+00:00","updated_at":"2026-06-25T01:18:48.013185+00:00"}