{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SQBGSVP3M2YBMM2UY37MEASUEX","short_pith_number":"pith:SQBGSVP3","schema_version":"1.0","canonical_sha256":"94026955fb66b0163354c6fec2025425eefecb04d35712d6b38631ad13191a6b","source":{"kind":"arxiv","id":"2502.03752","version":5},"attestation_state":"computed","paper":{"title":"Self-Improving Skill Learning for Robust Skill-based Meta-Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Sanghyeon Lee, Sangjun Bae, Seungyul Han, Yisak Park","submitted_at":"2025-02-06T03:28:45Z","abstract_excerpt":"Meta-reinforcement learning (Meta-RL) facilitates rapid adaptation to unseen tasks but faces challenges in long-horizon environments. Skill-based approaches tackle this by decomposing state-action sequences into reusable skills and employing hierarchical decision-making. However, these methods are highly susceptible to noisy offline demonstrations, leading to unstable skill learning and degraded performance. To address this, we propose Self-Improving Skill Learning (SISL), which performs self-guided skill refinement using decoupled high-level and skill improvement policies, while applying skil"},"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":"2502.03752","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-06T03:28:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"89aca2c0236f845d56ab8aea198d1fa2bf2f9fa2b55216dff17a3daecaee28dc","abstract_canon_sha256":"7e36fa0b20ee027c48120369f2c95a5feabaecc117cd1bb12740aabb54b4f631"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:12.854712Z","signature_b64":"c+DrktXJjdVr7fCKFcxc5OH1IyWcGys1BTOeN8CaJEHAciCaD244C7A68zALE9chyFBGMYacABF3ql6KCkKlCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94026955fb66b0163354c6fec2025425eefecb04d35712d6b38631ad13191a6b","last_reissued_at":"2026-05-21T01:04:12.853885Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:12.853885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Improving Skill Learning for Robust Skill-based Meta-Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Sanghyeon Lee, Sangjun Bae, Seungyul Han, Yisak Park","submitted_at":"2025-02-06T03:28:45Z","abstract_excerpt":"Meta-reinforcement learning (Meta-RL) facilitates rapid adaptation to unseen tasks but faces challenges in long-horizon environments. Skill-based approaches tackle this by decomposing state-action sequences into reusable skills and employing hierarchical decision-making. However, these methods are highly susceptible to noisy offline demonstrations, leading to unstable skill learning and degraded performance. To address this, we propose Self-Improving Skill Learning (SISL), which performs self-guided skill refinement using decoupled high-level and skill improvement policies, while applying skil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.03752","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/2502.03752/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":"2502.03752","created_at":"2026-05-21T01:04:12.854006+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03752v5","created_at":"2026-05-21T01:04:12.854006+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03752","created_at":"2026-05-21T01:04:12.854006+00:00"},{"alias_kind":"pith_short_12","alias_value":"SQBGSVP3M2YB","created_at":"2026-05-21T01:04:12.854006+00:00"},{"alias_kind":"pith_short_16","alias_value":"SQBGSVP3M2YBMM2U","created_at":"2026-05-21T01:04:12.854006+00:00"},{"alias_kind":"pith_short_8","alias_value":"SQBGSVP3","created_at":"2026-05-21T01:04:12.854006+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/SQBGSVP3M2YBMM2UY37MEASUEX","json":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX.json","graph_json":"https://pith.science/api/pith-number/SQBGSVP3M2YBMM2UY37MEASUEX/graph.json","events_json":"https://pith.science/api/pith-number/SQBGSVP3M2YBMM2UY37MEASUEX/events.json","paper":"https://pith.science/paper/SQBGSVP3"},"agent_actions":{"view_html":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX","download_json":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX.json","view_paper":"https://pith.science/paper/SQBGSVP3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03752&json=true","fetch_graph":"https://pith.science/api/pith-number/SQBGSVP3M2YBMM2UY37MEASUEX/graph.json","fetch_events":"https://pith.science/api/pith-number/SQBGSVP3M2YBMM2UY37MEASUEX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX/action/storage_attestation","attest_author":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX/action/author_attestation","sign_citation":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX/action/citation_signature","submit_replication":"https://pith.science/pith/SQBGSVP3M2YBMM2UY37MEASUEX/action/replication_record"}},"created_at":"2026-05-21T01:04:12.854006+00:00","updated_at":"2026-05-21T01:04:12.854006+00:00"}