{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RND6AARTW5VGVJBXVMQKBP7EAE","short_pith_number":"pith:RND6AART","canonical_record":{"source":{"id":"2506.02177","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T19:03:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"70002ab0940069bca99d1a2a4da6401eb1d0a531596b87ebff4d05de5c7a462c","abstract_canon_sha256":"59fa664d92ef60765eb31f1be3294fbc4c5ebb657074d459d3631020b70d083b"},"schema_version":"1.0"},"canonical_sha256":"8b47e00233b76a6aa437ab20a0bfe40108d2cf81a3618cded7c76f9a75ad4785","source":{"kind":"arxiv","id":"2506.02177","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02177","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02177v1","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02177","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_12","alias_value":"RND6AARTW5VG","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_16","alias_value":"RND6AARTW5VGVJBX","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_8","alias_value":"RND6AART","created_at":"2026-07-05T11:14:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RND6AARTW5VGVJBXVMQKBP7EAE","target":"record","payload":{"canonical_record":{"source":{"id":"2506.02177","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T19:03:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"70002ab0940069bca99d1a2a4da6401eb1d0a531596b87ebff4d05de5c7a462c","abstract_canon_sha256":"59fa664d92ef60765eb31f1be3294fbc4c5ebb657074d459d3631020b70d083b"},"schema_version":"1.0"},"canonical_sha256":"8b47e00233b76a6aa437ab20a0bfe40108d2cf81a3618cded7c76f9a75ad4785","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:33.771488Z","signature_b64":"b6Ucmu8yj7IH5DwMMrrQdaO5TDVcjQym5eHntiwR9F0WAPqimWIGehvmq1iYkqi4OuQhaXektoxrkap3MjQrAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b47e00233b76a6aa437ab20a0bfe40108d2cf81a3618cded7c76f9a75ad4785","last_reissued_at":"2026-07-05T11:14:33.770986Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:33.770986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.02177","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":"2026-07-05T11:14:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9m1vAOe/Vd1+W8DtKUjwn6YbpWeY5AvHXzshVLracit5rwitMNGxNlvcUpGe6HeP7lwN1LWJzgYgg0j/KOEoDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:52:30.596027Z"},"content_sha256":"385d660ed04623f8d9dd7e2231d0e84d2811581b907b56fe473847d1759e2620","schema_version":"1.0","event_id":"sha256:385d660ed04623f8d9dd7e2231d0e84d2811581b907b56fe473847d1759e2620"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RND6AARTW5VGVJBXVMQKBP7EAE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Beidi Chen, Bhavya Kailkhura, Brian R. Bartoldson, Fan Lai, Haizhong Zheng, Jiawei Zhao, Yang Zhou","submitted_at":"2025-06-02T19:03:00Z","abstract_excerpt":"Reinforcement learning, such as PPO and GRPO, has powered recent breakthroughs in LLM reasoning. Scaling rollout to sample more prompts enables models to selectively use higher-quality data for training, which can stabilize RL training and improve model performance. However, this comes at the cost of significant computational overhead. In this paper, we show that a substantial portion of this overhead can be avoided by skipping uninformative prompts before rollout. Our analysis of reward dynamics reveals a strong temporal consistency in prompt value: prompts that are uninformative in one epoch"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02177","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/2506.02177/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":"2026-07-05T11:14:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h7E8qgyB5mCCVyerDKdATl8/zdtu4WzJAwyfFt963lqXWS514to2k2pEQQEf1ph5DOeTJmIkDdA4O7MRYcg+DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:52:30.596568Z"},"content_sha256":"cf2874535b5341cce6d83fe958f22666c502c8a2fe33b1b19dd23bda48091d63","schema_version":"1.0","event_id":"sha256:cf2874535b5341cce6d83fe958f22666c502c8a2fe33b1b19dd23bda48091d63"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RND6AARTW5VGVJBXVMQKBP7EAE/bundle.json","state_url":"https://pith.science/pith/RND6AARTW5VGVJBXVMQKBP7EAE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RND6AARTW5VGVJBXVMQKBP7EAE/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":"2026-08-04T16:52:30Z","links":{"resolver":"https://pith.science/pith/RND6AARTW5VGVJBXVMQKBP7EAE","bundle":"https://pith.science/pith/RND6AARTW5VGVJBXVMQKBP7EAE/bundle.json","state":"https://pith.science/pith/RND6AARTW5VGVJBXVMQKBP7EAE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RND6AARTW5VGVJBXVMQKBP7EAE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RND6AARTW5VGVJBXVMQKBP7EAE","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":"59fa664d92ef60765eb31f1be3294fbc4c5ebb657074d459d3631020b70d083b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T19:03:00Z","title_canon_sha256":"70002ab0940069bca99d1a2a4da6401eb1d0a531596b87ebff4d05de5c7a462c"},"schema_version":"1.0","source":{"id":"2506.02177","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02177","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02177v1","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02177","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_12","alias_value":"RND6AARTW5VG","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_16","alias_value":"RND6AARTW5VGVJBX","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_8","alias_value":"RND6AART","created_at":"2026-07-05T11:14:33Z"}],"graph_snapshots":[{"event_id":"sha256:cf2874535b5341cce6d83fe958f22666c502c8a2fe33b1b19dd23bda48091d63","target":"graph","created_at":"2026-07-05T11:14:33Z","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/2506.02177/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning, such as PPO and GRPO, has powered recent breakthroughs in LLM reasoning. Scaling rollout to sample more prompts enables models to selectively use higher-quality data for training, which can stabilize RL training and improve model performance. However, this comes at the cost of significant computational overhead. In this paper, we show that a substantial portion of this overhead can be avoided by skipping uninformative prompts before rollout. Our analysis of reward dynamics reveals a strong temporal consistency in prompt value: prompts that are uninformative in one epoch","authors_text":"Beidi Chen, Bhavya Kailkhura, Brian R. Bartoldson, Fan Lai, Haizhong Zheng, Jiawei Zhao, Yang Zhou","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T19:03:00Z","title":"Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02177","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":"sha256:385d660ed04623f8d9dd7e2231d0e84d2811581b907b56fe473847d1759e2620","target":"record","created_at":"2026-07-05T11:14:33Z","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":"59fa664d92ef60765eb31f1be3294fbc4c5ebb657074d459d3631020b70d083b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T19:03:00Z","title_canon_sha256":"70002ab0940069bca99d1a2a4da6401eb1d0a531596b87ebff4d05de5c7a462c"},"schema_version":"1.0","source":{"id":"2506.02177","kind":"arxiv","version":1}},"canonical_sha256":"8b47e00233b76a6aa437ab20a0bfe40108d2cf81a3618cded7c76f9a75ad4785","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8b47e00233b76a6aa437ab20a0bfe40108d2cf81a3618cded7c76f9a75ad4785","first_computed_at":"2026-07-05T11:14:33.770986Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:33.770986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"b6Ucmu8yj7IH5DwMMrrQdaO5TDVcjQym5eHntiwR9F0WAPqimWIGehvmq1iYkqi4OuQhaXektoxrkap3MjQrAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:33.771488Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.02177","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:385d660ed04623f8d9dd7e2231d0e84d2811581b907b56fe473847d1759e2620","sha256:cf2874535b5341cce6d83fe958f22666c502c8a2fe33b1b19dd23bda48091d63"],"state_sha256":"095af999aca1e1150aebc85c62a546154fe4b362fc41463e65538d4d90625f94"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Cn0co9GtbDBe8rab5rCmIi/tvjgraVgNEFxofxABVxk57QaXKzrFpu0X/gcfd0v3uSvscv1/9qWhbAfRxXO3Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:52:30.602453Z","bundle_sha256":"2f7a834ee3b2014236c65e83f9072268829698b4afe6bb6d16a3b69d2547f8dc"}}