{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7JDQC4ZGP3FQRKSIQKNQ3EBS5G","short_pith_number":"pith:7JDQC4ZG","schema_version":"1.0","canonical_sha256":"fa470173267ecb08aa48829b0d9032e999a646c081765c078f8f53176ae5459b","source":{"kind":"arxiv","id":"2506.01442","version":2},"attestation_state":"computed","paper":{"title":"Agentic Episodic Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuyun Shen, Haosheng Chen, Junjie Sheng, Wenhao Li, Xiangfeng Wang, Xidong Yang, Yun Hua","submitted_at":"2025-06-02T08:57:37Z","abstract_excerpt":"Reinforcement learning (RL) remains fundamentally limited by poor data efficiency and weak generalization. Prior episodic RL methods attempt to alleviate this via external memory modules, yet they suffer from two key limitations: a representation bottleneck caused by shallow encoders, and a retrieval dilemma where episodic memory is accessed indiscriminately. To address these challenges, we propose Agentic Episodic Control (AEC), a novel architecture that integrates large language models (LLMs) into episodic RL. AEC uses an LLM-based semantic augmenter to generate semantic representations from"},"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":"2506.01442","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T08:57:37Z","cross_cats_sorted":[],"title_canon_sha256":"1504904a3ee135bad30ffa5defd2041d700e95b03fa19834485b7211d6320b8b","abstract_canon_sha256":"b2aebcf7a5e4eef2714b298a03b8e47fc3d89a732b17beda1d96af1cd8235d10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:14:02.693924Z","signature_b64":"oAuDhdM/6Fuh+OhVA0XqonSC3KmD1vmlrD8as8eeIa9GuiIRJh5ioQAgdr+M1Op/OKqF1V2Ia7dKUfhYhzXsAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa470173267ecb08aa48829b0d9032e999a646c081765c078f8f53176ae5459b","last_reissued_at":"2026-06-29T01:14:02.693434Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:14:02.693434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agentic Episodic Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuyun Shen, Haosheng Chen, Junjie Sheng, Wenhao Li, Xiangfeng Wang, Xidong Yang, Yun Hua","submitted_at":"2025-06-02T08:57:37Z","abstract_excerpt":"Reinforcement learning (RL) remains fundamentally limited by poor data efficiency and weak generalization. Prior episodic RL methods attempt to alleviate this via external memory modules, yet they suffer from two key limitations: a representation bottleneck caused by shallow encoders, and a retrieval dilemma where episodic memory is accessed indiscriminately. To address these challenges, we propose Agentic Episodic Control (AEC), a novel architecture that integrates large language models (LLMs) into episodic RL. AEC uses an LLM-based semantic augmenter to generate semantic representations from"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01442","kind":"arxiv","version":2},"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/2506.01442/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":"2506.01442","created_at":"2026-06-29T01:14:02.693492+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.01442v2","created_at":"2026-06-29T01:14:02.693492+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01442","created_at":"2026-06-29T01:14:02.693492+00:00"},{"alias_kind":"pith_short_12","alias_value":"7JDQC4ZGP3FQ","created_at":"2026-06-29T01:14:02.693492+00:00"},{"alias_kind":"pith_short_16","alias_value":"7JDQC4ZGP3FQRKSI","created_at":"2026-06-29T01:14:02.693492+00:00"},{"alias_kind":"pith_short_8","alias_value":"7JDQC4ZG","created_at":"2026-06-29T01:14:02.693492+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":268,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08468","citing_title":"PYTHALAB-MERA: Validation-Grounded Memory, Retrieval, and Acceptance Control for Frozen-LLM Coding Agents","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G","json":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G.json","graph_json":"https://pith.science/api/pith-number/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/graph.json","events_json":"https://pith.science/api/pith-number/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/events.json","paper":"https://pith.science/paper/7JDQC4ZG"},"agent_actions":{"view_html":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G","download_json":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G.json","view_paper":"https://pith.science/paper/7JDQC4ZG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.01442&json=true","fetch_graph":"https://pith.science/api/pith-number/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/graph.json","fetch_events":"https://pith.science/api/pith-number/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/action/storage_attestation","attest_author":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/action/author_attestation","sign_citation":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/action/citation_signature","submit_replication":"https://pith.science/pith/7JDQC4ZGP3FQRKSIQKNQ3EBS5G/action/replication_record"}},"created_at":"2026-06-29T01:14:02.693492+00:00","updated_at":"2026-06-29T01:14:02.693492+00:00"}