{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:3AMXKO7MXBTE3LAY5OED4B25UZ","short_pith_number":"pith:3AMXKO7M","canonical_record":{"source":{"id":"2602.17038","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-19T03:18:30Z","cross_cats_sorted":[],"title_canon_sha256":"b25dc7603a314881a13074dcfdcabe988450f4abaf5838e214c626872b8ab95f","abstract_canon_sha256":"bd90d5533239cf3b310dd4e5d37c50b2166543fa1b0f42b3b064fa74ed45697c"},"schema_version":"1.0"},"canonical_sha256":"d819753becb8664dac18eb883e075da644c1c3d955ac36931a73737e8176d407","source":{"kind":"arxiv","id":"2602.17038","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.17038","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"arxiv_version","alias_value":"2602.17038v3","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.17038","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_12","alias_value":"3AMXKO7MXBTE","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_16","alias_value":"3AMXKO7MXBTE3LAY","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_8","alias_value":"3AMXKO7M","created_at":"2026-05-20T01:05:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:3AMXKO7MXBTE3LAY5OED4B25UZ","target":"record","payload":{"canonical_record":{"source":{"id":"2602.17038","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-19T03:18:30Z","cross_cats_sorted":[],"title_canon_sha256":"b25dc7603a314881a13074dcfdcabe988450f4abaf5838e214c626872b8ab95f","abstract_canon_sha256":"bd90d5533239cf3b310dd4e5d37c50b2166543fa1b0f42b3b064fa74ed45697c"},"schema_version":"1.0"},"canonical_sha256":"d819753becb8664dac18eb883e075da644c1c3d955ac36931a73737e8176d407","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:09.624252Z","signature_b64":"oxNbXE3cto3hOjlrlsd5L0O/8V7fPPGjxi7/2CTOfNab6xjOj90KVz12EOQxmJWn9CRqaQ6sXzblmGr/Wn+zBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d819753becb8664dac18eb883e075da644c1c3d955ac36931a73737e8176d407","last_reissued_at":"2026-05-20T01:05:09.623423Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:09.623423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.17038","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T01:05:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1wIlqnWaEP4LQ7DfdTEmDyN1j2AQL58LjoNkiZmaQwA2eqO7KtHiH+qssGcdU7y/k4Hd2XPPFc93rHxscgpSDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:50:37.847650Z"},"content_sha256":"1db1c954869183057d57207046655101875dce9cd64e43d487672d7891ad0f31","schema_version":"1.0","event_id":"sha256:1db1c954869183057d57207046655101875dce9cd64e43d487672d7891ad0f31"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:3AMXKO7MXBTE3LAY5OED4B25UZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Phase-Aware Mixture of Experts for Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Lei Feng, Peng Jiang, Qingpeng Cai, Shengtian Yang, Shuo He, Yewen Li, Yu Li","submitted_at":"2026-02-19T03:18:30Z","abstract_excerpt":"Reinforcement learning (RL) has equipped LLM agents with a strong ability to solve complex tasks. However, existing RL methods normally use a \\emph{single} policy network, causing \\emph{simplicity bias} where simple tasks occupy most parameters and dominate gradient updates, leaving insufficient capacity for complex tasks. A plausible remedy could be employing the Mixture-of-Experts (MoE) architecture in the policy network, as MoE allows different parameters (experts) to specialize in different tasks, preventing simple tasks from dominating all parameters. However, a key limitation of traditio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.17038","kind":"arxiv","version":3},"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/2602.17038/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":"2026-05-20T01:05:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9qnJKfi+lR9Cq7egNPJ88f2+gyqmF0JRYwt5nfQWW7nk0Sn31a3uFREWnZa1689oHpUK4ld7FVuXMChu3C/lBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:50:37.848330Z"},"content_sha256":"6f9a40fb740f9fb1ec349060932fdc3e2b8b66aca5e68d4970129cb21cbc5d6c","schema_version":"1.0","event_id":"sha256:6f9a40fb740f9fb1ec349060932fdc3e2b8b66aca5e68d4970129cb21cbc5d6c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/bundle.json","state_url":"https://pith.science/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/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":"2026-05-26T06:50:37Z","links":{"resolver":"https://pith.science/pith/3AMXKO7MXBTE3LAY5OED4B25UZ","bundle":"https://pith.science/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/bundle.json","state":"https://pith.science/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3AMXKO7MXBTE3LAY5OED4B25UZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3AMXKO7MXBTE3LAY5OED4B25UZ","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":"bd90d5533239cf3b310dd4e5d37c50b2166543fa1b0f42b3b064fa74ed45697c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-19T03:18:30Z","title_canon_sha256":"b25dc7603a314881a13074dcfdcabe988450f4abaf5838e214c626872b8ab95f"},"schema_version":"1.0","source":{"id":"2602.17038","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.17038","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"arxiv_version","alias_value":"2602.17038v3","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.17038","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_12","alias_value":"3AMXKO7MXBTE","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_16","alias_value":"3AMXKO7MXBTE3LAY","created_at":"2026-05-20T01:05:09Z"},{"alias_kind":"pith_short_8","alias_value":"3AMXKO7M","created_at":"2026-05-20T01:05:09Z"}],"graph_snapshots":[{"event_id":"sha256:6f9a40fb740f9fb1ec349060932fdc3e2b8b66aca5e68d4970129cb21cbc5d6c","target":"graph","created_at":"2026-05-20T01:05:09Z","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/2602.17038/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) has equipped LLM agents with a strong ability to solve complex tasks. However, existing RL methods normally use a \\emph{single} policy network, causing \\emph{simplicity bias} where simple tasks occupy most parameters and dominate gradient updates, leaving insufficient capacity for complex tasks. A plausible remedy could be employing the Mixture-of-Experts (MoE) architecture in the policy network, as MoE allows different parameters (experts) to specialize in different tasks, preventing simple tasks from dominating all parameters. However, a key limitation of traditio","authors_text":"Lei Feng, Peng Jiang, Qingpeng Cai, Shengtian Yang, Shuo He, Yewen Li, Yu Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-19T03:18:30Z","title":"Phase-Aware Mixture of Experts for Agentic Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.17038","kind":"arxiv","version":3},"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":"sha256:1db1c954869183057d57207046655101875dce9cd64e43d487672d7891ad0f31","target":"record","created_at":"2026-05-20T01:05:09Z","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":"bd90d5533239cf3b310dd4e5d37c50b2166543fa1b0f42b3b064fa74ed45697c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-19T03:18:30Z","title_canon_sha256":"b25dc7603a314881a13074dcfdcabe988450f4abaf5838e214c626872b8ab95f"},"schema_version":"1.0","source":{"id":"2602.17038","kind":"arxiv","version":3}},"canonical_sha256":"d819753becb8664dac18eb883e075da644c1c3d955ac36931a73737e8176d407","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d819753becb8664dac18eb883e075da644c1c3d955ac36931a73737e8176d407","first_computed_at":"2026-05-20T01:05:09.623423Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:09.623423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oxNbXE3cto3hOjlrlsd5L0O/8V7fPPGjxi7/2CTOfNab6xjOj90KVz12EOQxmJWn9CRqaQ6sXzblmGr/Wn+zBA==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:09.624252Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.17038","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1db1c954869183057d57207046655101875dce9cd64e43d487672d7891ad0f31","sha256:6f9a40fb740f9fb1ec349060932fdc3e2b8b66aca5e68d4970129cb21cbc5d6c"],"state_sha256":"3917db3bcd9fc7778e94b0ea0cf64da713e31efe9b89b7145a4638404eba0d28"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RQgLjxSUoYyM5QIGxlDCWwTQPEEha18da74UXBkl25/utzn2j8tWrVPJIO1ObmvyF+sN3PHhoEKI8U/idB4iBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T06:50:37.852036Z","bundle_sha256":"074450822fa07aecc8f800ba2c72c760c713b50d8a551583f105b0e81f574a6b"}}