{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OCRY7AAYPDF2IIAX6UYZRIQ552","short_pith_number":"pith:OCRY7AAY","canonical_record":{"source":{"id":"2506.00439","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-31T07:38:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6befd0b8159f7e7ffecc099d1a6eb9bdc74076c769f6a22a75c2b12ae741e56f","abstract_canon_sha256":"0300172433a1e2d9db6455cf275893c506a4737b5634821fca37459052568392"},"schema_version":"1.0"},"canonical_sha256":"70a38f801878cba42017f53198a21deebe459497f64a99d28aa98a91871f5901","source":{"kind":"arxiv","id":"2506.00439","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00439","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00439v1","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00439","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_12","alias_value":"OCRY7AAYPDF2","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_16","alias_value":"OCRY7AAYPDF2IIAX","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_8","alias_value":"OCRY7AAY","created_at":"2026-07-05T11:13:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OCRY7AAYPDF2IIAX6UYZRIQ552","target":"record","payload":{"canonical_record":{"source":{"id":"2506.00439","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-31T07:38:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6befd0b8159f7e7ffecc099d1a6eb9bdc74076c769f6a22a75c2b12ae741e56f","abstract_canon_sha256":"0300172433a1e2d9db6455cf275893c506a4737b5634821fca37459052568392"},"schema_version":"1.0"},"canonical_sha256":"70a38f801878cba42017f53198a21deebe459497f64a99d28aa98a91871f5901","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:34.773890Z","signature_b64":"K74uZddEZNEQ3YSgn02XXzT/ZgrvVe7FysRf9qDtmlbX1HDwdHejAVnULsIFnZJBtPSXkqd9EjnDAGe7GqNJBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70a38f801878cba42017f53198a21deebe459497f64a99d28aa98a91871f5901","last_reissued_at":"2026-07-05T11:13:34.773363Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:34.773363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.00439","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":"2026-07-05T11:13:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M1C7KuYVRcooIrd+KvxloHJUh3iKt8z+G4GNH5I5aLCWUb44+JRbOIJWDm5D30Psg3xqA52Xg/JzLncs7YgyBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T13:56:31.292119Z"},"content_sha256":"fa515464482cfe9cc25f55551fed363570d37316514d487195f268913569f132","schema_version":"1.0","event_id":"sha256:fa515464482cfe9cc25f55551fed363570d37316514d487195f268913569f132"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OCRY7AAYPDF2IIAX6UYZRIQ552","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RLAE: Reinforcement Learning-Assisted Ensemble for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongbin Zhao, Guojun Yin, Jiajun Chai, Qichao Zhang, Wei Lin, Yuanheng Zhu, Yuqian Fu","submitted_at":"2025-05-31T07:38:41Z","abstract_excerpt":"Ensembling large language models (LLMs) can effectively combine diverse strengths of different models, offering a promising approach to enhance performance across various tasks. However, existing methods typically rely on fixed weighting strategies that fail to adapt to the dynamic, context-dependent characteristics of LLM capabilities. In this work, we propose Reinforcement Learning-Assisted Ensemble for LLMs (RLAE), a novel framework that reformulates LLM ensemble through the lens of a Markov Decision Process (MDP). Our approach introduces a RL agent that dynamically adjusts ensemble weights"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00439","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/2506.00439/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":"2026-07-05T11:13:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0JPa8Hyf4cjuqFmKF5qRpqgt9BGlw9IS15zCEBJ+42IqKZDiKfeDkIxzHRuwVrTsULhB7KOJybu4aLjTlS+qDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T13:56:31.292512Z"},"content_sha256":"a3285f56ac8fd8e88f9b358e1a8f6c76a47e06d43909befed41ebeeb7d1f5c0a","schema_version":"1.0","event_id":"sha256:a3285f56ac8fd8e88f9b358e1a8f6c76a47e06d43909befed41ebeeb7d1f5c0a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/bundle.json","state_url":"https://pith.science/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/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":"2026-07-12T13:56:31Z","links":{"resolver":"https://pith.science/pith/OCRY7AAYPDF2IIAX6UYZRIQ552","bundle":"https://pith.science/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/bundle.json","state":"https://pith.science/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OCRY7AAYPDF2IIAX6UYZRIQ552/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OCRY7AAYPDF2IIAX6UYZRIQ552","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":"0300172433a1e2d9db6455cf275893c506a4737b5634821fca37459052568392","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-31T07:38:41Z","title_canon_sha256":"6befd0b8159f7e7ffecc099d1a6eb9bdc74076c769f6a22a75c2b12ae741e56f"},"schema_version":"1.0","source":{"id":"2506.00439","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00439","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00439v1","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00439","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_12","alias_value":"OCRY7AAYPDF2","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_16","alias_value":"OCRY7AAYPDF2IIAX","created_at":"2026-07-05T11:13:34Z"},{"alias_kind":"pith_short_8","alias_value":"OCRY7AAY","created_at":"2026-07-05T11:13:34Z"}],"graph_snapshots":[{"event_id":"sha256:a3285f56ac8fd8e88f9b358e1a8f6c76a47e06d43909befed41ebeeb7d1f5c0a","target":"graph","created_at":"2026-07-05T11:13:34Z","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/2506.00439/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Ensembling large language models (LLMs) can effectively combine diverse strengths of different models, offering a promising approach to enhance performance across various tasks. However, existing methods typically rely on fixed weighting strategies that fail to adapt to the dynamic, context-dependent characteristics of LLM capabilities. In this work, we propose Reinforcement Learning-Assisted Ensemble for LLMs (RLAE), a novel framework that reformulates LLM ensemble through the lens of a Markov Decision Process (MDP). Our approach introduces a RL agent that dynamically adjusts ensemble weights","authors_text":"Dongbin Zhao, Guojun Yin, Jiajun Chai, Qichao Zhang, Wei Lin, Yuanheng Zhu, Yuqian Fu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-31T07:38:41Z","title":"RLAE: Reinforcement Learning-Assisted Ensemble for LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00439","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:fa515464482cfe9cc25f55551fed363570d37316514d487195f268913569f132","target":"record","created_at":"2026-07-05T11:13:34Z","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":"0300172433a1e2d9db6455cf275893c506a4737b5634821fca37459052568392","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-31T07:38:41Z","title_canon_sha256":"6befd0b8159f7e7ffecc099d1a6eb9bdc74076c769f6a22a75c2b12ae741e56f"},"schema_version":"1.0","source":{"id":"2506.00439","kind":"arxiv","version":1}},"canonical_sha256":"70a38f801878cba42017f53198a21deebe459497f64a99d28aa98a91871f5901","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"70a38f801878cba42017f53198a21deebe459497f64a99d28aa98a91871f5901","first_computed_at":"2026-07-05T11:13:34.773363Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:34.773363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"K74uZddEZNEQ3YSgn02XXzT/ZgrvVe7FysRf9qDtmlbX1HDwdHejAVnULsIFnZJBtPSXkqd9EjnDAGe7GqNJBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:34.773890Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.00439","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fa515464482cfe9cc25f55551fed363570d37316514d487195f268913569f132","sha256:a3285f56ac8fd8e88f9b358e1a8f6c76a47e06d43909befed41ebeeb7d1f5c0a"],"state_sha256":"c7fd1b04f0641d74460e4b50d7b915230d24a063a255ff3406bec458ba88388b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UqKn+ekfTPPpEaWWnyzuVfFPu0F1NiDg7ClKBNCey+DEzRjc/S/8qNOFpvP4f0zKpoUjXi+0bDEEE8g9OeuiCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T13:56:31.294729Z","bundle_sha256":"b1538ba8c0e196eabbcb05983f6db0480ec48d1f11ca7ef6aace0ca028cf9dd3"}}