{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KEOBPYJSC7DJ34WIRQCD6XKORF","short_pith_number":"pith:KEOBPYJS","schema_version":"1.0","canonical_sha256":"511c17e13217c69df2c88c043f5d4e89761bc350d29a835e5617c3e4655b3048","source":{"kind":"arxiv","id":"2505.16997","version":1},"attestation_state":"computed","paper":{"title":"X-MAS: Towards Building Multi-Agent Systems with Heterogeneous LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.MA"],"primary_cat":"cs.AI","authors_text":"Lei Bai, Qimin Wu, Rui Ye, Siheng Chen, Xianghe Pang, Xiangrui Liu, Zhenfei Yin","submitted_at":"2025-05-22T17:56:39Z","abstract_excerpt":"LLM-based multi-agent systems (MAS) extend the capabilities of single LLMs by enabling cooperation among multiple specialized agents. However, most existing MAS frameworks rely on a single LLM to drive all agents, constraining the system's intelligence to the limit of that model. This paper explores the paradigm of heterogeneous LLM-driven MAS (X-MAS), where agents are powered by diverse LLMs, elevating the system's potential to the collective intelligence of diverse LLMs. We introduce X-MAS-Bench, a comprehensive testbed designed to evaluate the performance of various LLMs across different do"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.16997","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-22T17:56:39Z","cross_cats_sorted":["cs.CL","cs.MA"],"title_canon_sha256":"bfd9756050c612ba5053623b524fd6a42bceffe0a6ebc0a5c4502f24d7f6cfe7","abstract_canon_sha256":"9a8287362681ac4e93fd8e50eb8bab5a00ad8be76b35766826614e64e908f956"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:44.293203Z","signature_b64":"ucRdS/EEL0y7AeEKPj55uwODQwTtG3ZinUJPFvPx5FGhmelmW+sTQcZyDoLEGCAg5lLus4VB7aflcOYdt7E6AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"511c17e13217c69df2c88c043f5d4e89761bc350d29a835e5617c3e4655b3048","last_reissued_at":"2026-07-05T11:07:44.292703Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:44.292703Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"X-MAS: Towards Building Multi-Agent Systems with Heterogeneous LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.MA"],"primary_cat":"cs.AI","authors_text":"Lei Bai, Qimin Wu, Rui Ye, Siheng Chen, Xianghe Pang, Xiangrui Liu, Zhenfei Yin","submitted_at":"2025-05-22T17:56:39Z","abstract_excerpt":"LLM-based multi-agent systems (MAS) extend the capabilities of single LLMs by enabling cooperation among multiple specialized agents. However, most existing MAS frameworks rely on a single LLM to drive all agents, constraining the system's intelligence to the limit of that model. This paper explores the paradigm of heterogeneous LLM-driven MAS (X-MAS), where agents are powered by diverse LLMs, elevating the system's potential to the collective intelligence of diverse LLMs. We introduce X-MAS-Bench, a comprehensive testbed designed to evaluate the performance of various LLMs across different do"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16997","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/2505.16997/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.16997","created_at":"2026-07-05T11:07:44.292763+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16997v1","created_at":"2026-07-05T11:07:44.292763+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16997","created_at":"2026-07-05T11:07:44.292763+00:00"},{"alias_kind":"pith_short_12","alias_value":"KEOBPYJSC7DJ","created_at":"2026-07-05T11:07:44.292763+00:00"},{"alias_kind":"pith_short_16","alias_value":"KEOBPYJSC7DJ34WI","created_at":"2026-07-05T11:07:44.292763+00:00"},{"alias_kind":"pith_short_8","alias_value":"KEOBPYJS","created_at":"2026-07-05T11:07:44.292763+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24428","citing_title":"Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26112","citing_title":"From Model Scaling to System Scaling: Scaling the Harness in Agentic AI","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15573","citing_title":"Response-Conditioned Parallel-to-Sequential Orchestration for Multi-Agent Systems","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":173,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12024","citing_title":"Beyond Sentiment: A Multi-Agent Pipeline for Actionable Business Advice from Reviews","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":172,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21046","citing_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","ref_index":269,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13647","citing_title":"FlowCompile: An Optimizing Compiler for Structured LLM Workflows","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22446","citing_title":"From Skills to Talent: Organising Heterogeneous Agents as a Real-World Company","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27351","citing_title":"Heterogeneous Scientific Foundation Model Collaboration","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF","json":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF.json","graph_json":"https://pith.science/api/pith-number/KEOBPYJSC7DJ34WIRQCD6XKORF/graph.json","events_json":"https://pith.science/api/pith-number/KEOBPYJSC7DJ34WIRQCD6XKORF/events.json","paper":"https://pith.science/paper/KEOBPYJS"},"agent_actions":{"view_html":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF","download_json":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF.json","view_paper":"https://pith.science/paper/KEOBPYJS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16997&json=true","fetch_graph":"https://pith.science/api/pith-number/KEOBPYJSC7DJ34WIRQCD6XKORF/graph.json","fetch_events":"https://pith.science/api/pith-number/KEOBPYJSC7DJ34WIRQCD6XKORF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF/action/storage_attestation","attest_author":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF/action/author_attestation","sign_citation":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF/action/citation_signature","submit_replication":"https://pith.science/pith/KEOBPYJSC7DJ34WIRQCD6XKORF/action/replication_record"}},"created_at":"2026-07-05T11:07:44.292763+00:00","updated_at":"2026-07-05T11:07:44.292763+00:00"}