{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:67G7AXIDSFBUU6NXZQZ6O4SQDH","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":"af776880cb8ce2c3e4806d9cfd2fe45edaa416f8a331f745c8443fa6169eb135","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T22:55:18Z","title_canon_sha256":"e0a11feb5d8b1cd8d5371e64ca145ef1936b3d87c36083152e984c159318b28e"},"schema_version":"1.0","source":{"id":"2606.04272","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.04272","created_at":"2026-06-04T01:09:01Z"},{"alias_kind":"arxiv_version","alias_value":"2606.04272v1","created_at":"2026-06-04T01:09:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04272","created_at":"2026-06-04T01:09:01Z"},{"alias_kind":"pith_short_12","alias_value":"67G7AXIDSFBU","created_at":"2026-06-04T01:09:01Z"},{"alias_kind":"pith_short_16","alias_value":"67G7AXIDSFBUU6NX","created_at":"2026-06-04T01:09:01Z"},{"alias_kind":"pith_short_8","alias_value":"67G7AXID","created_at":"2026-06-04T01:09:01Z"}],"graph_snapshots":[{"event_id":"sha256:0f1e0b1ae074889a3dcbea747eb21e1b3711884e591aa640163ff883a21663b3","target":"graph","created_at":"2026-06-04T01:09:01Z","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/2606.04272/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The standard LLM training pipeline applies reinforcement learning (RL) only after pre-training and supervised fine-tuning (SFT). We question this status quo by training a LLM from scratch and applying RL, SFT, and SFT followed by RL directly to intermediate pre-training checkpoints. We find that RL is effective very early, and often matches the full SFT$\\to$RL pipeline early as well. Through experiments on harder problems, we find that targeted pre-training data composition is a strong lever for RL effectiveness, even more so than model scale. Beyond reasoning accuracy, applying RL directly to","authors_text":"Clara Mohri, David Alvarez-Melis, Rachit Bansal, Sham Kakade, Tian Qin","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T22:55:18Z","title":"RL Excursions during Pre-Training: Re-examining Policy Optimization for LLM training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04272","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":"sha256:7f922807c12e35e98a8c0c4e6bd041d9bcb5ff4ecb6054cd6cc34fc99a74e6e7","target":"record","created_at":"2026-06-04T01:09:01Z","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":"af776880cb8ce2c3e4806d9cfd2fe45edaa416f8a331f745c8443fa6169eb135","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T22:55:18Z","title_canon_sha256":"e0a11feb5d8b1cd8d5371e64ca145ef1936b3d87c36083152e984c159318b28e"},"schema_version":"1.0","source":{"id":"2606.04272","kind":"arxiv","version":1}},"canonical_sha256":"f7cdf05d0391434a79b7cc33e7725019d95409c0b344a67b917c5e2ea8e1cdaa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7cdf05d0391434a79b7cc33e7725019d95409c0b344a67b917c5e2ea8e1cdaa","first_computed_at":"2026-06-04T01:09:01.200946Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:09:01.200946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p0SQkLnV9sNV/oiN8HPcY8E3oI084IQ9cCTH3vmxTkNWl17NvdhTQQdtK1Bk9OX5OowCnvD+yK1okWaHrUJrCg==","signature_status":"signed_v1","signed_at":"2026-06-04T01:09:01.201399Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.04272","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7f922807c12e35e98a8c0c4e6bd041d9bcb5ff4ecb6054cd6cc34fc99a74e6e7","sha256:0f1e0b1ae074889a3dcbea747eb21e1b3711884e591aa640163ff883a21663b3"],"state_sha256":"5cb93e283c606893f12244089ea8238afca00b778560be20f430fe1b54c4bfbf"}