{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TYFQR5V6WHUWY7MEVY4BI5UOMH","short_pith_number":"pith:TYFQR5V6","schema_version":"1.0","canonical_sha256":"9e0b08f6beb1e96c7d84ae3814768e61d3a02edba71c2468a6fec879ce5bc29f","source":{"kind":"arxiv","id":"2604.07799","version":2},"attestation_state":"computed","paper":{"title":"Learning Without Losing Identity: Capability Evolution for Embodied Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Embodied agents improve task success from 32% to 91% by evolving separate capability modules without altering their core identity or safety limits.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Cong Yang, John See, Simin Luan, Xue Qin, Zhijun Li","submitted_at":"2026-04-09T04:51:07Z","abstract_excerpt":"Embodied agents are expected to operate persistently in dynamic physical environments, continuously acquiring new capabilities over time. Existing approaches to improving agent performance often rely on modifying the agent itself -- through prompt engineering, policy updates, or structural redesign -- leading to instability and loss of identity in long-lived systems. In this work, we propose a capability-centric evolution paradigm for embodied agents. We argue that a robot should maintain a persistent agent as its cognitive identity, while enabling continuous improvement through the evolution "},"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":"2604.07799","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-09T04:51:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3edf72d21b37be9a9c8d7643d80b8ae9c709a7060d82aa981eb6286ba6f8a37e","abstract_canon_sha256":"aa37fea55ebea829264347c29b1a406e5cc205bb475d8e924e771860c3ee6b40"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:01.807283Z","signature_b64":"27shYs8bUNz2YH+ifz0TQIvDlqnYV+p4lmndmEutn2r7HA8I5DUDUadcglj7YguVgVe+dYEAWuKBuIbum5UBAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e0b08f6beb1e96c7d84ae3814768e61d3a02edba71c2468a6fec879ce5bc29f","last_reissued_at":"2026-05-22T01:04:01.806262Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:01.806262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Without Losing Identity: Capability Evolution for Embodied Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Embodied agents improve task success from 32% to 91% by evolving separate capability modules without altering their core identity or safety limits.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Cong Yang, John See, Simin Luan, Xue Qin, Zhijun Li","submitted_at":"2026-04-09T04:51:07Z","abstract_excerpt":"Embodied agents are expected to operate persistently in dynamic physical environments, continuously acquiring new capabilities over time. Existing approaches to improving agent performance often rely on modifying the agent itself -- through prompt engineering, policy updates, or structural redesign -- leading to instability and loss of identity in long-lived systems. In this work, we propose a capability-centric evolution paradigm for embodied agents. We argue that a robot should maintain a persistent agent as its cognitive identity, while enabling continuous improvement through the evolution "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We demonstrate through simulated embodied tasks that capability evolution improves task success rates from 32.4% to 91.3% over 20 iterations, outperforming both agent-modification baselines and established skill-learning methods (SPiRL, SkiMo), while preserving zero policy drift and zero safety violations.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That embodied capabilities can be cleanly decomposed into independent, versioned ECMs whose evolution leaves the persistent agent identity and safety constraints untouched.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Embodied agents maintain a persistent identity while evolving capabilities via modular ECMs, raising simulated task success from 32.4% to 91.3% over 20 iterations with zero policy drift or safety violations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Embodied agents improve task success from 32% to 91% by evolving separate capability modules without altering their core identity or safety limits.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"025bce8f149746d01a17d1f66eaeb298cd6927cbe705085d74d200b95cc9f585"},"source":{"id":"2604.07799","kind":"arxiv","version":2},"verdict":{"id":"2b3a9046-31dd-4f3a-b9fa-e8c8f559aa4c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T18:05:59.766335Z","strongest_claim":"We demonstrate through simulated embodied tasks that capability evolution improves task success rates from 32.4% to 91.3% over 20 iterations, outperforming both agent-modification baselines and established skill-learning methods (SPiRL, SkiMo), while preserving zero policy drift and zero safety violations.","one_line_summary":"Embodied agents maintain a persistent identity while evolving capabilities via modular ECMs, raising simulated task success from 32.4% to 91.3% over 20 iterations with zero policy drift or safety violations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That embodied capabilities can be cleanly decomposed into independent, versioned ECMs whose evolution leaves the persistent agent identity and safety constraints untouched.","pith_extraction_headline":"Embodied agents improve task success from 32% to 91% by evolving separate capability modules without altering their core identity or safety limits."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.07799/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":"2604.07799","created_at":"2026-05-22T01:04:01.806391+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.07799v2","created_at":"2026-05-22T01:04:01.806391+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.07799","created_at":"2026-05-22T01:04:01.806391+00:00"},{"alias_kind":"pith_short_12","alias_value":"TYFQR5V6WHUW","created_at":"2026-05-22T01:04:01.806391+00:00"},{"alias_kind":"pith_short_16","alias_value":"TYFQR5V6WHUWY7ME","created_at":"2026-05-22T01:04:01.806391+00:00"},{"alias_kind":"pith_short_8","alias_value":"TYFQR5V6","created_at":"2026-05-22T01:04:01.806391+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2604.07833","citing_title":"Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11028","citing_title":"Federated Single-Agent Robotics: Multi-Robot Coordination Without Intra-Robot Multi-Agent Fragmentation","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11174","citing_title":"EmbodiedGovBench: A Benchmark for Governance, Recovery, and Upgrade Safety in Embodied Agent Systems","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11028","citing_title":"Federated Single-Agent Robotics: Multi-Robot Coordination Without Intra-Robot Multi-Agent Fragmentation","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2604.13097","citing_title":"ECM Contracts: Contract-Aware, Versioned, and Governable Capability Interfaces for Embodied Agents","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07833","citing_title":"Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution","ref_index":60,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH","json":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH.json","graph_json":"https://pith.science/api/pith-number/TYFQR5V6WHUWY7MEVY4BI5UOMH/graph.json","events_json":"https://pith.science/api/pith-number/TYFQR5V6WHUWY7MEVY4BI5UOMH/events.json","paper":"https://pith.science/paper/TYFQR5V6"},"agent_actions":{"view_html":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH","download_json":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH.json","view_paper":"https://pith.science/paper/TYFQR5V6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.07799&json=true","fetch_graph":"https://pith.science/api/pith-number/TYFQR5V6WHUWY7MEVY4BI5UOMH/graph.json","fetch_events":"https://pith.science/api/pith-number/TYFQR5V6WHUWY7MEVY4BI5UOMH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH/action/storage_attestation","attest_author":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH/action/author_attestation","sign_citation":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH/action/citation_signature","submit_replication":"https://pith.science/pith/TYFQR5V6WHUWY7MEVY4BI5UOMH/action/replication_record"}},"created_at":"2026-05-22T01:04:01.806391+00:00","updated_at":"2026-05-22T01:04:01.806391+00:00"}