{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PXNTP5ANF7T3XLKGS4BHBA3UFZ","short_pith_number":"pith:PXNTP5AN","schema_version":"1.0","canonical_sha256":"7ddb37f40d2fe7bbad4697027083742e55d4e7b1b9c997b215ff22d3c2f48e0d","source":{"kind":"arxiv","id":"2603.12056","version":3},"attestation_state":"computed","paper":{"title":"XSkill: Continual Learning from Experience and Skills in Multimodal Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guanyu Jiang, Xiaoye Qu, Yi R. Fung, Zhaochen Su","submitted_at":"2026-03-12T15:25:57Z","abstract_excerpt":"Multimodal agents can now tackle complex reasoning tasks with diverse tools, yet they still suffer from inefficient tool use and inflexible orchestration in open-ended settings. A central challenge is enabling such agents to continually improve without parameter updates by learning from past trajectories. We identify two complementary forms of reusable knowledge essential for this goal: experiences, providing concise action-level guidance for tool selection and decision making, and skills, providing structured task-level guidance for planning and tool use. To this end, we propose XSkill, a dua"},"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":"2603.12056","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T15:25:57Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ecddd14f238ed9d0b6071ccd839357e76ebc484d6a876c6b6dbc4379531c5f18","abstract_canon_sha256":"e766293af78735b42b5e3586f7ef0e1049d6e966b26fd2d69d0a039e97396f46"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T01:17:29.364048Z","signature_b64":"Cti+N2q3GdD9WWp7zLqb1VJjQEsr/M+xZ0m/ORF9eShsKEItb3fytDg2O1SyWhLdy8eACdTnb6lIV4utxnf8DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ddb37f40d2fe7bbad4697027083742e55d4e7b1b9c997b215ff22d3c2f48e0d","last_reissued_at":"2026-07-02T01:17:29.363532Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T01:17:29.363532Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"XSkill: Continual Learning from Experience and Skills in Multimodal Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guanyu Jiang, Xiaoye Qu, Yi R. Fung, Zhaochen Su","submitted_at":"2026-03-12T15:25:57Z","abstract_excerpt":"Multimodal agents can now tackle complex reasoning tasks with diverse tools, yet they still suffer from inefficient tool use and inflexible orchestration in open-ended settings. A central challenge is enabling such agents to continually improve without parameter updates by learning from past trajectories. We identify two complementary forms of reusable knowledge essential for this goal: experiences, providing concise action-level guidance for tool selection and decision making, and skills, providing structured task-level guidance for planning and tool use. To this end, we propose XSkill, a dua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.12056","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/2603.12056/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":"2603.12056","created_at":"2026-07-02T01:17:29.363592+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.12056v3","created_at":"2026-07-02T01:17:29.363592+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.12056","created_at":"2026-07-02T01:17:29.363592+00:00"},{"alias_kind":"pith_short_12","alias_value":"PXNTP5ANF7T3","created_at":"2026-07-02T01:17:29.363592+00:00"},{"alias_kind":"pith_short_16","alias_value":"PXNTP5ANF7T3XLKG","created_at":"2026-07-02T01:17:29.363592+00:00"},{"alias_kind":"pith_short_8","alias_value":"PXNTP5AN","created_at":"2026-07-02T01:17:29.363592+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":24,"internal_anchor_count":24,"sample":[{"citing_arxiv_id":"2606.03056","citing_title":"SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.26200","citing_title":"Workflow Closure Is Not Scientific Closure in Auto-Research Systems","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":97,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13527","citing_title":"MMSkills: Towards Multimodal Skills for General Visual Agents","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07549","citing_title":"PathoSage: Towards Multi-Source Evidence Adjudication in Pathology via Experience-Aware Agentic Workflow","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20631","citing_title":"Harnessing Agent Skills: Architectural Patterns and a Reference Architecture for Skill-Mediated LLM Agents","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01139","citing_title":"SkillRevise: Improving LLM-Authored Agent Skills via Trace-Conditioned Skill Revision","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01314","citing_title":"SkillSmith: Co-Evolving Skills and Tools for Self-Improving Agent Systems","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":103,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16363","citing_title":"ORACLE: Anticipating Scams from Partial Trajectories in Streaming App Usage","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17933","citing_title":"AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13527","citing_title":"MMSkills: Towards Multimodal Skills for General Visual Agents","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13527","citing_title":"MMSkills: Towards Multimodal Skills for General Visual Agents","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.04017","citing_title":"GeoBrowse: A Geolocation Benchmark for Agentic Tool Use with Expert-Annotated Reasoning Traces","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11169","citing_title":"OLIVIA: Online Learning via Inference-time Action Adaptation for Decision Making in LLM ReAct Agents","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08526","citing_title":"Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10332","citing_title":"EmbodiSkill: Skill-Aware Reflection for Self-Evolving Embodied Agents","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05726","citing_title":"SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":103,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08043","citing_title":"SCOPE: Structured Decomposition and Conditional Skill Orchestration for Complex Image Generation","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18292","citing_title":"Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ","json":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ.json","graph_json":"https://pith.science/api/pith-number/PXNTP5ANF7T3XLKGS4BHBA3UFZ/graph.json","events_json":"https://pith.science/api/pith-number/PXNTP5ANF7T3XLKGS4BHBA3UFZ/events.json","paper":"https://pith.science/paper/PXNTP5AN"},"agent_actions":{"view_html":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ","download_json":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ.json","view_paper":"https://pith.science/paper/PXNTP5AN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.12056&json=true","fetch_graph":"https://pith.science/api/pith-number/PXNTP5ANF7T3XLKGS4BHBA3UFZ/graph.json","fetch_events":"https://pith.science/api/pith-number/PXNTP5ANF7T3XLKGS4BHBA3UFZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ/action/storage_attestation","attest_author":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ/action/author_attestation","sign_citation":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ/action/citation_signature","submit_replication":"https://pith.science/pith/PXNTP5ANF7T3XLKGS4BHBA3UFZ/action/replication_record"}},"created_at":"2026-07-02T01:17:29.363592+00:00","updated_at":"2026-07-02T01:17:29.363592+00:00"}