{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6OYASPGP4U3NAIZNNYFQ7JOCUD","short_pith_number":"pith:6OYASPGP","schema_version":"1.0","canonical_sha256":"f3b0093ccfe536d0232d6e0b0fa5c2a0fe949dd12c3c32301dd0b6dc78baf6c3","source":{"kind":"arxiv","id":"2506.14990","version":3},"attestation_state":"computed","paper":{"title":"MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Andreas Bulling, Bram Grooten, Constantin Ruhdorfer, Fabrice Kusters, Luka van den Boogaard, Meng Fang, Mykola Pechenizkiy, Samuel Garcin, Tristan Tomilin, Yali Du","submitted_at":"2025-06-17T21:50:04Z","abstract_excerpt":"Benchmarks play a central role in reinforcement learning (RL) research, yet their computational constraints often shape what is studied. Despite the motivation of lifelong learning, most continual RL papers consider only 3-10 sequential tasks, as CPU-bound environments make longer sequences impractical. Meanwhile, continual learning in cooperative multi-agent settings remains largely unexplored. To address these gaps, we introduce MEAL (Multi-agent Environments for Adaptive Learning), the first benchmark for continual multi-agent RL. By leveraging JAX and GPU acceleration, MEAL enables trainin"},"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":"2506.14990","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-17T21:50:04Z","cross_cats_sorted":[],"title_canon_sha256":"df657ceef357a47a9e6496fd19cd4a009260bb30e30503b62837084443e2e8a4","abstract_canon_sha256":"3805b52cb56f713b2607bb6958d9b22b32f5d0550913b59e3b3c31dcfedc059c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:46.286473Z","signature_b64":"LY/jfVVPcmjnq0GgYE2Yt1nVBZVWUA7FzuAXGy/vo+9ulZSPRcokNlUpfclkWF3/16mVkn6glYsfLL/htORlBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f3b0093ccfe536d0232d6e0b0fa5c2a0fe949dd12c3c32301dd0b6dc78baf6c3","last_reissued_at":"2026-06-19T16:12:46.285352Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:46.285352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Andreas Bulling, Bram Grooten, Constantin Ruhdorfer, Fabrice Kusters, Luka van den Boogaard, Meng Fang, Mykola Pechenizkiy, Samuel Garcin, Tristan Tomilin, Yali Du","submitted_at":"2025-06-17T21:50:04Z","abstract_excerpt":"Benchmarks play a central role in reinforcement learning (RL) research, yet their computational constraints often shape what is studied. Despite the motivation of lifelong learning, most continual RL papers consider only 3-10 sequential tasks, as CPU-bound environments make longer sequences impractical. Meanwhile, continual learning in cooperative multi-agent settings remains largely unexplored. To address these gaps, we introduce MEAL (Multi-agent Environments for Adaptive Learning), the first benchmark for continual multi-agent RL. By leveraging JAX and GPU acceleration, MEAL enables trainin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14990","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/2506.14990/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":"2506.14990","created_at":"2026-06-19T16:12:46.285930+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.14990v3","created_at":"2026-06-19T16:12:46.285930+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14990","created_at":"2026-06-19T16:12:46.285930+00:00"},{"alias_kind":"pith_short_12","alias_value":"6OYASPGP4U3N","created_at":"2026-06-19T16:12:46.285930+00:00"},{"alias_kind":"pith_short_16","alias_value":"6OYASPGP4U3NAIZN","created_at":"2026-06-19T16:12:46.285930+00:00"},{"alias_kind":"pith_short_8","alias_value":"6OYASPGP","created_at":"2026-06-19T16:12:46.285930+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2511.20857","citing_title":"Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory","ref_index":102,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD","json":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD.json","graph_json":"https://pith.science/api/pith-number/6OYASPGP4U3NAIZNNYFQ7JOCUD/graph.json","events_json":"https://pith.science/api/pith-number/6OYASPGP4U3NAIZNNYFQ7JOCUD/events.json","paper":"https://pith.science/paper/6OYASPGP"},"agent_actions":{"view_html":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD","download_json":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD.json","view_paper":"https://pith.science/paper/6OYASPGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.14990&json=true","fetch_graph":"https://pith.science/api/pith-number/6OYASPGP4U3NAIZNNYFQ7JOCUD/graph.json","fetch_events":"https://pith.science/api/pith-number/6OYASPGP4U3NAIZNNYFQ7JOCUD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD/action/storage_attestation","attest_author":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD/action/author_attestation","sign_citation":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD/action/citation_signature","submit_replication":"https://pith.science/pith/6OYASPGP4U3NAIZNNYFQ7JOCUD/action/replication_record"}},"created_at":"2026-06-19T16:12:46.285930+00:00","updated_at":"2026-06-19T16:12:46.285930+00:00"}