{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PPVKZDEHDDS4UEBGUQD3IP2TPF","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":"90b5dd0ee5f8c0671ad18c3c6869262239d1d5d063f971c2ea82535446863831","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:11:27Z","title_canon_sha256":"2a3ceed12129c862013a513d84303942bd8ca228ef545f36479cf01b494d748f"},"schema_version":"1.0","source":{"id":"2605.14212","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14212","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14212v1","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14212","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"pith_short_12","alias_value":"PPVKZDEHDDS4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"PPVKZDEHDDS4UEBG","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"PPVKZDEH","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:70d188f319453a331f41e517f3067c916437379bc8fc253dd374038b42a9e0f0","target":"graph","created_at":"2026-05-17T23:39:10Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"MetaAgent-X consistently outperforms existing automatic MAS baselines, achieving up to 21.7% gains. ... These results establish end-to-end trainable automatic MAS as a practical paradigm for building self-designing and self-executing agentic models."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That Executor Designer Hierarchical Rollout and Stagewise Co-evolution provide stable joint optimization and accurate credit assignment across designer and executor trajectories without introducing new instabilities or biases that would prevent both components from improving."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"MetaAgent-X uses end-to-end RL to jointly optimize automatic multi-agent system design and execution, outperforming baselines by up to 21.7% through hierarchical rollouts and stagewise co-evolution."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"MetaAgent-X jointly trains the designer and executors of automatic multi-agent systems using end-to-end reinforcement learning."}],"snapshot_sha256":"76b7e384b93f58cdc485ece5cef168a0be915c86185b3c43061d562b113e25f2"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8db17f618145b6f3b5bbc4d3213a0fe744253cf312ed4f399402b985c038bd42"},"paper":{"abstract_excerpt":"Automatic multi-agent systems aim to instantiate agent workflows without relying on manually designed or fixed orchestration. However, existing automatic MAS approaches remain only partially adaptive: they either perform training-free test-time search or optimize the meta-level designer while keeping downstream execution agents frozen, which creating a frozen-executor ceiling and leaving the end-to-end training of self-designing and self-executing agentic models unexplored. To address this, we introduce MetaAgent-X, an end-to-end reinforcement learning framework that jointly optimizes automati","authors_text":"Huazheng Wang, Jiayu Chang, Jishen Zhao, Nan Wang, Qingyun Wu, Yaolun Zhang, Yiran Wu, Yizhao Chen, Yujie Zhao","cross_cats":[],"headline":"MetaAgent-X jointly trains the designer and executors of automatic multi-agent systems using end-to-end reinforcement learning.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:11:27Z","title":"MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learning"},"references":{"count":52,"internal_anchors":0,"resolved_work":52,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Yujie Zhao, Hejia Zhang, Hanxian Huang, Zhongming Yu, and Jishen Zhao","work_id":"e6d87867-de15-4827-b740-f4aa603f708a","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"14 Figure 4: Sensitivity analysis on the stage length for designer–executor alternation","work_id":"3590c28c-4ac0-41c9-b839-a869f25bb253","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Delivery formatting.Inter-agent messages must be strictly enclosed within <delivery>...</delivery> tags. This constraint serves a dual purpose: it establishes a structured, easily parsable communicati","work_id":"cadd2d18-128e-43b1-918c-7001c4cce933","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Compute total number of ways to choose 4 numbers from 10:C(10,4)","work_id":"2aa90731-24bf-47c6-9ffd-29d736a4d47c","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"ComputeP(grand prize): number of ways to match all 4 numbers","work_id":"df92995a-93fa-4a13-8312-47293fb49251","year":null}],"snapshot_sha256":"d34860785c889846c0b0a127339043d0ffd501cd70baa0e7b2bc9fd0e614884a"},"source":{"id":"2605.14212","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:45:37.586097Z","id":"3bde2c6f-5fdf-4a2f-b951-47ddd30f87c8","model_set":{"reader":"grok-4.3"},"one_line_summary":"MetaAgent-X uses end-to-end RL to jointly optimize automatic multi-agent system design and execution, outperforming baselines by up to 21.7% through hierarchical rollouts and stagewise co-evolution.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"MetaAgent-X jointly trains the designer and executors of automatic multi-agent systems using end-to-end reinforcement learning.","strongest_claim":"MetaAgent-X consistently outperforms existing automatic MAS baselines, achieving up to 21.7% gains. ... These results establish end-to-end trainable automatic MAS as a practical paradigm for building self-designing and self-executing agentic models.","weakest_assumption":"That Executor Designer Hierarchical Rollout and Stagewise Co-evolution provide stable joint optimization and accurate credit assignment across designer and executor trajectories without introducing new instabilities or biases that would prevent both components from improving."}},"verdict_id":"3bde2c6f-5fdf-4a2f-b951-47ddd30f87c8"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e7fe106bd8b3b7336c252c3507594632048c0399cd7ee747c3dc254739d69e98","target":"record","created_at":"2026-05-17T23:39:10Z","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":"90b5dd0ee5f8c0671ad18c3c6869262239d1d5d063f971c2ea82535446863831","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:11:27Z","title_canon_sha256":"2a3ceed12129c862013a513d84303942bd8ca228ef545f36479cf01b494d748f"},"schema_version":"1.0","source":{"id":"2605.14212","kind":"arxiv","version":1}},"canonical_sha256":"7beaac8c8718e5ca1026a407b43f53795a72a8ee1a859b555d1376ab95508428","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7beaac8c8718e5ca1026a407b43f53795a72a8ee1a859b555d1376ab95508428","first_computed_at":"2026-05-17T23:39:10.919082Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:10.919082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HK+mz49jBjAmPmiE4/+2lxXeLalHw7bfsBjsUS7dflYZxX+mJRWix2uZ0HcIqpjZrgD+3FsMARL6fX6QOAFPAw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:10.919510Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14212","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e7fe106bd8b3b7336c252c3507594632048c0399cd7ee747c3dc254739d69e98","sha256:70d188f319453a331f41e517f3067c916437379bc8fc253dd374038b42a9e0f0"],"state_sha256":"5c74e06319ea9bcf6ca0eec7f44ce7f89787944d931e86f8268cbde844675fc0"}