{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JWFUXZ7G6QGQXANMJQGT2FP73N","short_pith_number":"pith:JWFUXZ7G","schema_version":"1.0","canonical_sha256":"4d8b4be7e6f40d0b81ac4c0d3d15ffdb7c4ab1c37ff6dac76768bc8fe2257857","source":{"kind":"arxiv","id":"2410.01706","version":5},"attestation_state":"computed","paper":{"title":"Sable: a Performant, Efficient and Scalable Sequence Model for MARL","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Arnu Pretorius, Claude Formanek, Jemma Daniel, Liam Clark, Louay Ben Nessir, Louise Beyers, Omayma Mahjoub, Ruan de Kock, Sasha Abramowitz, Simon du Toit, Wiem Khlifi","submitted_at":"2024-10-02T16:15:26Z","abstract_excerpt":"As multi-agent reinforcement learning (MARL) progresses towards solving larger and more complex problems, it becomes increasingly important that algorithms exhibit the key properties of (1) strong performance, (2) memory efficiency, and (3) scalability. In this work, we introduce Sable, a performant, memory-efficient, and scalable sequence modeling approach to MARL. Sable works by adapting the retention mechanism in Retentive Networks (Sun et al., 2023) to achieve computationally efficient processing of multi-agent observations with long context memory for temporal reasoning. Through extensive"},"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":"2410.01706","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-02T16:15:26Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"f72667983bbb03ee19ec03f934da8e7adb13be1cf99aff03f90553a396e03fa3","abstract_canon_sha256":"38c8a0bea5e2ceecd2db3e8fef631f52b9b0f47e04c633789b6afb413391bcd3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:43.450760Z","signature_b64":"lIYFP7VjMZqywgIBlZFeRW/cHx6IJeOm8XH5GWLuyqBhYjrqSzMZtiZv9k1jC5KELSgbO76QIcrt2dRSaFlUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d8b4be7e6f40d0b81ac4c0d3d15ffdb7c4ab1c37ff6dac76768bc8fe2257857","last_reissued_at":"2026-07-05T11:09:43.450222Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:43.450222Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sable: a Performant, Efficient and Scalable Sequence Model for MARL","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Arnu Pretorius, Claude Formanek, Jemma Daniel, Liam Clark, Louay Ben Nessir, Louise Beyers, Omayma Mahjoub, Ruan de Kock, Sasha Abramowitz, Simon du Toit, Wiem Khlifi","submitted_at":"2024-10-02T16:15:26Z","abstract_excerpt":"As multi-agent reinforcement learning (MARL) progresses towards solving larger and more complex problems, it becomes increasingly important that algorithms exhibit the key properties of (1) strong performance, (2) memory efficiency, and (3) scalability. In this work, we introduce Sable, a performant, memory-efficient, and scalable sequence modeling approach to MARL. Sable works by adapting the retention mechanism in Retentive Networks (Sun et al., 2023) to achieve computationally efficient processing of multi-agent observations with long context memory for temporal reasoning. Through extensive"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01706","kind":"arxiv","version":5},"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/2410.01706/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":"2410.01706","created_at":"2026-07-05T11:09:43.450288+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01706v5","created_at":"2026-07-05T11:09:43.450288+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01706","created_at":"2026-07-05T11:09:43.450288+00:00"},{"alias_kind":"pith_short_12","alias_value":"JWFUXZ7G6QGQ","created_at":"2026-07-05T11:09:43.450288+00:00"},{"alias_kind":"pith_short_16","alias_value":"JWFUXZ7G6QGQXANM","created_at":"2026-07-05T11:09:43.450288+00:00"},{"alias_kind":"pith_short_8","alias_value":"JWFUXZ7G","created_at":"2026-07-05T11:09:43.450288+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.13554","citing_title":"Self-Supervised On-Policy Reinforcement Learning via Contrastive Proximal Policy Optimisation","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06825","citing_title":"Randomness is sometimes necessary for coordination","ref_index":83,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N","json":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N.json","graph_json":"https://pith.science/api/pith-number/JWFUXZ7G6QGQXANMJQGT2FP73N/graph.json","events_json":"https://pith.science/api/pith-number/JWFUXZ7G6QGQXANMJQGT2FP73N/events.json","paper":"https://pith.science/paper/JWFUXZ7G"},"agent_actions":{"view_html":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N","download_json":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N.json","view_paper":"https://pith.science/paper/JWFUXZ7G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01706&json=true","fetch_graph":"https://pith.science/api/pith-number/JWFUXZ7G6QGQXANMJQGT2FP73N/graph.json","fetch_events":"https://pith.science/api/pith-number/JWFUXZ7G6QGQXANMJQGT2FP73N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N/action/storage_attestation","attest_author":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N/action/author_attestation","sign_citation":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N/action/citation_signature","submit_replication":"https://pith.science/pith/JWFUXZ7G6QGQXANMJQGT2FP73N/action/replication_record"}},"created_at":"2026-07-05T11:09:43.450288+00:00","updated_at":"2026-07-05T11:09:43.450288+00:00"}