{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NJTGNH2WCVRYVDQ22CFUDN5MYP","short_pith_number":"pith:NJTGNH2W","canonical_record":{"source":{"id":"2606.03073","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:02:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3219ad8d6dbe1f6db8678de5c25c03cdd782e80f5202cabf1d089c3928f3c7c4","abstract_canon_sha256":"b94972e59d97f57b25ddc6a9d9ebcc3a626f83d8cbc7cc13bd2c61f47dd51078"},"schema_version":"1.0"},"canonical_sha256":"6a66669f5615638a8e1ad08b41b7acc3dc904c4a6c76a448142070efb7593a88","source":{"kind":"arxiv","id":"2606.03073","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.03073","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"arxiv_version","alias_value":"2606.03073v1","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03073","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_12","alias_value":"NJTGNH2WCVRY","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_16","alias_value":"NJTGNH2WCVRYVDQ2","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_8","alias_value":"NJTGNH2W","created_at":"2026-06-03T01:05:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NJTGNH2WCVRYVDQ22CFUDN5MYP","target":"record","payload":{"canonical_record":{"source":{"id":"2606.03073","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:02:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3219ad8d6dbe1f6db8678de5c25c03cdd782e80f5202cabf1d089c3928f3c7c4","abstract_canon_sha256":"b94972e59d97f57b25ddc6a9d9ebcc3a626f83d8cbc7cc13bd2c61f47dd51078"},"schema_version":"1.0"},"canonical_sha256":"6a66669f5615638a8e1ad08b41b7acc3dc904c4a6c76a448142070efb7593a88","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:30.912622Z","signature_b64":"tbo7wSBiHg0bsL44Mpdi0VGiYToDtG/N2Y4urOTUxv8//hEBvJy4f69Eegg8/M9CG6kFqCEFNbIwkjJyXVoSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a66669f5615638a8e1ad08b41b7acc3dc904c4a6c76a448142070efb7593a88","last_reissued_at":"2026-06-03T01:05:30.912239Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:30.912239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.03073","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":"2026-06-03T01:05:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JjILiWaotwL1JxhrN0KqHQh6BG+l1rKz8zoSdmWsCUepg83ctRiBqQ3CTS8i+cc2rhNOrkXtbsEBGCMrUlJkDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-20T20:21:49.582543Z"},"content_sha256":"e50ccc79d6a75ee1490cb0fb9afe4e7fbdebea7a64736e4646eece103eba849f","schema_version":"1.0","event_id":"sha256:e50ccc79d6a75ee1490cb0fb9afe4e7fbdebea7a64736e4646eece103eba849f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NJTGNH2WCVRYVDQ22CFUDN5MYP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Hyperparameter Optimization for LLM Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bowen Xiao, Chuxuan Zeng, Du liang, Minping Chen, Zeyi Wen","submitted_at":"2026-06-02T03:02:06Z","abstract_excerpt":"Reinforcement learning (RL) for large language models (LLMs) is highly sensitive to hyperparameter configurations, making hyperparameter optimization (HPO) essential yet computationally expensive. Existing multi-fidelity HPO methods remain inefficient for LLM RL due to the massive model scale and resource-intensive training cycles. In this paper, we propose Joint Fidelity Hyperparameter Optimization (JF-HPO), which simultaneously adapts both model size and training budget as fidelity. JF-HPO is empowered by: (i) it leverages a small proxy model of the target LLM for efficient training and eval"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03073","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/2606.03073/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":"2026-06-03T01:05:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+58Hny89Wgzy4KN5Z0vlZ0XkAmfT6fUppNET407qfB2LB1eCrg9aucOp7122aHpA/Lkl1QzN3rSrgRjyIVuZCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-20T20:21:49.582926Z"},"content_sha256":"d83dc2c5fe0eceed0d112229efb92da7ffbc16748f3c94062abd3d3dd51a2476","schema_version":"1.0","event_id":"sha256:d83dc2c5fe0eceed0d112229efb92da7ffbc16748f3c94062abd3d3dd51a2476"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/bundle.json","state_url":"https://pith.science/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/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":"2026-06-20T20:21:49Z","links":{"resolver":"https://pith.science/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP","bundle":"https://pith.science/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/bundle.json","state":"https://pith.science/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NJTGNH2WCVRYVDQ22CFUDN5MYP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NJTGNH2WCVRYVDQ22CFUDN5MYP","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":"b94972e59d97f57b25ddc6a9d9ebcc3a626f83d8cbc7cc13bd2c61f47dd51078","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:02:06Z","title_canon_sha256":"3219ad8d6dbe1f6db8678de5c25c03cdd782e80f5202cabf1d089c3928f3c7c4"},"schema_version":"1.0","source":{"id":"2606.03073","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.03073","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"arxiv_version","alias_value":"2606.03073v1","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03073","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_12","alias_value":"NJTGNH2WCVRY","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_16","alias_value":"NJTGNH2WCVRYVDQ2","created_at":"2026-06-03T01:05:30Z"},{"alias_kind":"pith_short_8","alias_value":"NJTGNH2W","created_at":"2026-06-03T01:05:30Z"}],"graph_snapshots":[{"event_id":"sha256:d83dc2c5fe0eceed0d112229efb92da7ffbc16748f3c94062abd3d3dd51a2476","target":"graph","created_at":"2026-06-03T01:05:30Z","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/2606.03073/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) for large language models (LLMs) is highly sensitive to hyperparameter configurations, making hyperparameter optimization (HPO) essential yet computationally expensive. Existing multi-fidelity HPO methods remain inefficient for LLM RL due to the massive model scale and resource-intensive training cycles. In this paper, we propose Joint Fidelity Hyperparameter Optimization (JF-HPO), which simultaneously adapts both model size and training budget as fidelity. JF-HPO is empowered by: (i) it leverages a small proxy model of the target LLM for efficient training and eval","authors_text":"Bowen Xiao, Chuxuan Zeng, Du liang, Minping Chen, Zeyi Wen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:02:06Z","title":"Efficient Hyperparameter Optimization for LLM Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03073","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:e50ccc79d6a75ee1490cb0fb9afe4e7fbdebea7a64736e4646eece103eba849f","target":"record","created_at":"2026-06-03T01:05:30Z","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":"b94972e59d97f57b25ddc6a9d9ebcc3a626f83d8cbc7cc13bd2c61f47dd51078","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:02:06Z","title_canon_sha256":"3219ad8d6dbe1f6db8678de5c25c03cdd782e80f5202cabf1d089c3928f3c7c4"},"schema_version":"1.0","source":{"id":"2606.03073","kind":"arxiv","version":1}},"canonical_sha256":"6a66669f5615638a8e1ad08b41b7acc3dc904c4a6c76a448142070efb7593a88","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6a66669f5615638a8e1ad08b41b7acc3dc904c4a6c76a448142070efb7593a88","first_computed_at":"2026-06-03T01:05:30.912239Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-03T01:05:30.912239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tbo7wSBiHg0bsL44Mpdi0VGiYToDtG/N2Y4urOTUxv8//hEBvJy4f69Eegg8/M9CG6kFqCEFNbIwkjJyXVoSBQ==","signature_status":"signed_v1","signed_at":"2026-06-03T01:05:30.912622Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.03073","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e50ccc79d6a75ee1490cb0fb9afe4e7fbdebea7a64736e4646eece103eba849f","sha256:d83dc2c5fe0eceed0d112229efb92da7ffbc16748f3c94062abd3d3dd51a2476"],"state_sha256":"40973bea9f519c794b4dfdfce94221b6d99b3689343202628789aea3a8f2c152"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nKjKSaoeODwmwgMiVZV2SIsRo7+ctXIMQgPm3FMtG2tWKifud+EQuCFOZrUUQOBl96REqE/wJRQ2zMnidViLAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-20T20:21:49.584928Z","bundle_sha256":"3f87c2ab73ffa3b48efe6f1d895acf3866931ec1e26566ef503b22f9df6ede7d"}}