{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MJAC5SGJOL6FVX5X7LJXN5ZKPA","short_pith_number":"pith:MJAC5SGJ","schema_version":"1.0","canonical_sha256":"62402ec8c972fc5adfb7fad376f72a783bf6a308a7afff32d43cfdd1067c44b4","source":{"kind":"arxiv","id":"2505.11100","version":1},"attestation_state":"computed","paper":{"title":"Bidirectional Distillation: A Mixed-Play Framework for Multi-Agent Generalizable Behaviors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"De Ma, Dong Xing, Gang Pan, Jiahao Lin, Lang Feng, Li Zhang","submitted_at":"2025-05-16T10:31:10Z","abstract_excerpt":"Population-population generalization is a challenging problem in multi-agent reinforcement learning (MARL), particularly when agents encounter unseen co-players. However, existing self-play-based methods are constrained by the limitation of inside-space generalization. In this study, we propose Bidirectional Distillation (BiDist), a novel mixed-play framework, to overcome this limitation in MARL. BiDist leverages knowledge distillation in two alternating directions: forward distillation, which emulates the historical policies' space and creates an implicit self-play, and reverse distillation, "},"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":"2505.11100","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T10:31:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5925b5c33f8483d507f4b5d1670f9ad78f976d577ec9dd21a67489a39a0cf037","abstract_canon_sha256":"e2a7e5beab2831864ad097fe42b1acb7f3b6ab337f82c53ffd4c2383795ed6c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:10.784671Z","signature_b64":"Mm+wGM5zlD6fsblllEwHHxcVv/xU0qanvrJ7IFCPnBJN2IZKG3MvJA9ATdCv2pfvgnVbh/f2Lyvg+TOPi14HCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62402ec8c972fc5adfb7fad376f72a783bf6a308a7afff32d43cfdd1067c44b4","last_reissued_at":"2026-07-05T11:04:10.784196Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:10.784196Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bidirectional Distillation: A Mixed-Play Framework for Multi-Agent Generalizable Behaviors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"De Ma, Dong Xing, Gang Pan, Jiahao Lin, Lang Feng, Li Zhang","submitted_at":"2025-05-16T10:31:10Z","abstract_excerpt":"Population-population generalization is a challenging problem in multi-agent reinforcement learning (MARL), particularly when agents encounter unseen co-players. However, existing self-play-based methods are constrained by the limitation of inside-space generalization. In this study, we propose Bidirectional Distillation (BiDist), a novel mixed-play framework, to overcome this limitation in MARL. BiDist leverages knowledge distillation in two alternating directions: forward distillation, which emulates the historical policies' space and creates an implicit self-play, and reverse distillation, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11100","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/2505.11100/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":"2505.11100","created_at":"2026-07-05T11:04:10.784260+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11100v1","created_at":"2026-07-05T11:04:10.784260+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11100","created_at":"2026-07-05T11:04:10.784260+00:00"},{"alias_kind":"pith_short_12","alias_value":"MJAC5SGJOL6F","created_at":"2026-07-05T11:04:10.784260+00:00"},{"alias_kind":"pith_short_16","alias_value":"MJAC5SGJOL6FVX5X","created_at":"2026-07-05T11:04:10.784260+00:00"},{"alias_kind":"pith_short_8","alias_value":"MJAC5SGJ","created_at":"2026-07-05T11:04:10.784260+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA","json":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA.json","graph_json":"https://pith.science/api/pith-number/MJAC5SGJOL6FVX5X7LJXN5ZKPA/graph.json","events_json":"https://pith.science/api/pith-number/MJAC5SGJOL6FVX5X7LJXN5ZKPA/events.json","paper":"https://pith.science/paper/MJAC5SGJ"},"agent_actions":{"view_html":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA","download_json":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA.json","view_paper":"https://pith.science/paper/MJAC5SGJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11100&json=true","fetch_graph":"https://pith.science/api/pith-number/MJAC5SGJOL6FVX5X7LJXN5ZKPA/graph.json","fetch_events":"https://pith.science/api/pith-number/MJAC5SGJOL6FVX5X7LJXN5ZKPA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA/action/storage_attestation","attest_author":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA/action/author_attestation","sign_citation":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA/action/citation_signature","submit_replication":"https://pith.science/pith/MJAC5SGJOL6FVX5X7LJXN5ZKPA/action/replication_record"}},"created_at":"2026-07-05T11:04:10.784260+00:00","updated_at":"2026-07-05T11:04:10.784260+00:00"}