{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:N4OHBJPZUY3YQXWDKELNP7KX64","short_pith_number":"pith:N4OHBJPZ","canonical_record":{"source":{"id":"2605.21214","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:08:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"463cd73e4f41213d55cb9207b0f40840340b94c91c36ffade5317772de819aa3","abstract_canon_sha256":"9fa00dd8f47d808de6661c1d6c84f7cc691db1e577e588a0baf658927b06f677"},"schema_version":"1.0"},"canonical_sha256":"6f1c70a5f9a637885ec35116d7fd57f72bc82804ddf5fdcf984d500605c97683","source":{"kind":"arxiv","id":"2605.21214","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21214","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21214v1","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21214","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_12","alias_value":"N4OHBJPZUY3Y","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_16","alias_value":"N4OHBJPZUY3YQXWD","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_8","alias_value":"N4OHBJPZ","created_at":"2026-05-21T01:05:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:N4OHBJPZUY3YQXWDKELNP7KX64","target":"record","payload":{"canonical_record":{"source":{"id":"2605.21214","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:08:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"463cd73e4f41213d55cb9207b0f40840340b94c91c36ffade5317772de819aa3","abstract_canon_sha256":"9fa00dd8f47d808de6661c1d6c84f7cc691db1e577e588a0baf658927b06f677"},"schema_version":"1.0"},"canonical_sha256":"6f1c70a5f9a637885ec35116d7fd57f72bc82804ddf5fdcf984d500605c97683","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:43.882914Z","signature_b64":"/hNWy6p7niAR0ectSIYEIlyH4Sd2g7bNjcBz8PebBSpARbuWQ0xJ6HJcmDD0yhbnMdPEzLH2nrDQVnhbSkCHCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f1c70a5f9a637885ec35116d7fd57f72bc82804ddf5fdcf984d500605c97683","last_reissued_at":"2026-05-21T01:05:43.882271Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:43.882271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.21214","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":"2026-05-21T01:05:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YVkby0vzA6d17LxFCsGXrwvGj1tKpeBrx2ejVEKtdGgn1CUF8KiivoE+96CyCj3qgYFBftGb3xvKU/g237YaDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T08:49:23.159567Z"},"content_sha256":"65e18006f9a29cac6626154e33e081614eaa34cc558b8ad18fabe87373131e1c","schema_version":"1.0","event_id":"sha256:65e18006f9a29cac6626154e33e081614eaa34cc558b8ad18fabe87373131e1c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:N4OHBJPZUY3YQXWDKELNP7KX64","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Behavior-Consistent Deep Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benjamin Eysenbach, Claas Voelcker, Eric Eaton, Liv G. d'Aliberti, Marcel Hussing","submitted_at":"2026-05-20T14:08:33Z","abstract_excerpt":"Reinforcement learning (RL) often exhibits high variance across training runs, leading to unreliable performance and posing a major challenge to deployment in real-world domains. In this work, we address the challenge of cross-run policy divergence by formalizing the problem of behavior-consistent RL, where the objective is to obtain policies that are both high-performing and distributionally similar across training runs. Our key observation is that maximum-entropy RL provides a direct mechanism for controlling behavioral divergence by anchoring runs to a common (uniform) prior. We prove that,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21214","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/2605.21214/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":"2026-05-21T01:05:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ch90/a8fUFRdh1Jqjx02znlVSfiiOKXpMb0cppyP373dDVtdreik4nBgwnHDDuk5m+sxLIJjFB3t2mg9GKntAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T08:49:23.159948Z"},"content_sha256":"53b2131077d885e5b0418e1384946135a63310f0adc35997baf304c146ca8edc","schema_version":"1.0","event_id":"sha256:53b2131077d885e5b0418e1384946135a63310f0adc35997baf304c146ca8edc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N4OHBJPZUY3YQXWDKELNP7KX64/bundle.json","state_url":"https://pith.science/pith/N4OHBJPZUY3YQXWDKELNP7KX64/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N4OHBJPZUY3YQXWDKELNP7KX64/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":"2026-05-28T08:49:23Z","links":{"resolver":"https://pith.science/pith/N4OHBJPZUY3YQXWDKELNP7KX64","bundle":"https://pith.science/pith/N4OHBJPZUY3YQXWDKELNP7KX64/bundle.json","state":"https://pith.science/pith/N4OHBJPZUY3YQXWDKELNP7KX64/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N4OHBJPZUY3YQXWDKELNP7KX64/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:N4OHBJPZUY3YQXWDKELNP7KX64","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":"9fa00dd8f47d808de6661c1d6c84f7cc691db1e577e588a0baf658927b06f677","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:08:33Z","title_canon_sha256":"463cd73e4f41213d55cb9207b0f40840340b94c91c36ffade5317772de819aa3"},"schema_version":"1.0","source":{"id":"2605.21214","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21214","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21214v1","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21214","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_12","alias_value":"N4OHBJPZUY3Y","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_16","alias_value":"N4OHBJPZUY3YQXWD","created_at":"2026-05-21T01:05:43Z"},{"alias_kind":"pith_short_8","alias_value":"N4OHBJPZ","created_at":"2026-05-21T01:05:43Z"}],"graph_snapshots":[{"event_id":"sha256:53b2131077d885e5b0418e1384946135a63310f0adc35997baf304c146ca8edc","target":"graph","created_at":"2026-05-21T01:05:43Z","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/2605.21214/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) often exhibits high variance across training runs, leading to unreliable performance and posing a major challenge to deployment in real-world domains. In this work, we address the challenge of cross-run policy divergence by formalizing the problem of behavior-consistent RL, where the objective is to obtain policies that are both high-performing and distributionally similar across training runs. Our key observation is that maximum-entropy RL provides a direct mechanism for controlling behavioral divergence by anchoring runs to a common (uniform) prior. We prove that,","authors_text":"Benjamin Eysenbach, Claas Voelcker, Eric Eaton, Liv G. d'Aliberti, Marcel Hussing","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:08:33Z","title":"Behavior-Consistent Deep Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21214","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":"sha256:65e18006f9a29cac6626154e33e081614eaa34cc558b8ad18fabe87373131e1c","target":"record","created_at":"2026-05-21T01:05:43Z","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":"9fa00dd8f47d808de6661c1d6c84f7cc691db1e577e588a0baf658927b06f677","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:08:33Z","title_canon_sha256":"463cd73e4f41213d55cb9207b0f40840340b94c91c36ffade5317772de819aa3"},"schema_version":"1.0","source":{"id":"2605.21214","kind":"arxiv","version":1}},"canonical_sha256":"6f1c70a5f9a637885ec35116d7fd57f72bc82804ddf5fdcf984d500605c97683","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6f1c70a5f9a637885ec35116d7fd57f72bc82804ddf5fdcf984d500605c97683","first_computed_at":"2026-05-21T01:05:43.882271Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:05:43.882271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/hNWy6p7niAR0ectSIYEIlyH4Sd2g7bNjcBz8PebBSpARbuWQ0xJ6HJcmDD0yhbnMdPEzLH2nrDQVnhbSkCHCA==","signature_status":"signed_v1","signed_at":"2026-05-21T01:05:43.882914Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.21214","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:65e18006f9a29cac6626154e33e081614eaa34cc558b8ad18fabe87373131e1c","sha256:53b2131077d885e5b0418e1384946135a63310f0adc35997baf304c146ca8edc"],"state_sha256":"7ab34b0f3e739a02705d72e0ef6d03d6463168306c9e5e890960b47f761bf0d5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"glxqr0U9u5Bq3lbP3W8RaLXse7JrxHltrUKQTalInXmQ8YpmCiFxU1VRjVYpyYhUq/SkXTzQKQnMY8osWN8JAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T08:49:23.162069Z","bundle_sha256":"7d7ae004a8c8c8ba7853e46957172cecc24e7f65cfe53014150d84d2a6d98003"}}