{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WJFUAE6BJU3BEVAVFZHKE5N5YB","short_pith_number":"pith:WJFUAE6B","schema_version":"1.0","canonical_sha256":"b24b4013c14d361254152e4ea275bdc077b684a5931848a1fc66b7da743d0916","source":{"kind":"arxiv","id":"2505.17716","version":1},"attestation_state":"computed","paper":{"title":"Get Experience from Practice: LLM Agents with Record & Replay","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.LG","authors_text":"Cheng Zhang, Dong Du, Erhu Feng, Haibo Chen, Le Chen, Wenbo Zhou, Yisheng Zhao, Yubin Xia, Yunpeng Dong, Zhichao Hua, Zibin Liu","submitted_at":"2025-05-23T10:33:14Z","abstract_excerpt":"AI agents, empowered by Large Language Models (LLMs) and communication protocols such as MCP and A2A, have rapidly evolved from simple chatbots to autonomous entities capable of executing complex, multi-step tasks, demonstrating great potential. However, the LLMs' inherent uncertainty and heavy computational resource requirements pose four significant challenges to the development of safe and efficient agents: reliability, privacy, cost and performance. Existing approaches, like model alignment, workflow constraints and on-device model deployment, can partially alleviate some issues but often "},"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":"2505.17716","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T10:33:14Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"0346c5df62ff04f1b625f074a9dcd32a2a5e107435ad79ac1a26a17238222361","abstract_canon_sha256":"f0f65bc66df2758b559813e97b56b72cf35a677f0ad03ac3f2c684098a98cea6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:30.782074Z","signature_b64":"E+CEqym6/+YAltRRGtoVNGkUAviXXUdXrbBxVRuh1olr6Bzza97Fh2xXST4pAvk6gv8HupT6CwD7Gy2BkMKPBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b24b4013c14d361254152e4ea275bdc077b684a5931848a1fc66b7da743d0916","last_reissued_at":"2026-07-05T11:08:30.781546Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:30.781546Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Get Experience from Practice: LLM Agents with Record & Replay","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.LG","authors_text":"Cheng Zhang, Dong Du, Erhu Feng, Haibo Chen, Le Chen, Wenbo Zhou, Yisheng Zhao, Yubin Xia, Yunpeng Dong, Zhichao Hua, Zibin Liu","submitted_at":"2025-05-23T10:33:14Z","abstract_excerpt":"AI agents, empowered by Large Language Models (LLMs) and communication protocols such as MCP and A2A, have rapidly evolved from simple chatbots to autonomous entities capable of executing complex, multi-step tasks, demonstrating great potential. However, the LLMs' inherent uncertainty and heavy computational resource requirements pose four significant challenges to the development of safe and efficient agents: reliability, privacy, cost and performance. Existing approaches, like model alignment, workflow constraints and on-device model deployment, can partially alleviate some issues but often "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17716","kind":"arxiv","version":1},"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/2505.17716/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":"2505.17716","created_at":"2026-07-05T11:08:30.781608+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.17716v1","created_at":"2026-07-05T11:08:30.781608+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17716","created_at":"2026-07-05T11:08:30.781608+00:00"},{"alias_kind":"pith_short_12","alias_value":"WJFUAE6BJU3B","created_at":"2026-07-05T11:08:30.781608+00:00"},{"alias_kind":"pith_short_16","alias_value":"WJFUAE6BJU3BEVAV","created_at":"2026-07-05T11:08:30.781608+00:00"},{"alias_kind":"pith_short_8","alias_value":"WJFUAE6B","created_at":"2026-07-05T11:08:30.781608+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17929","citing_title":"PreAct: Computer-Using Agents that Get Faster on Repeated Tasks","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":264,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29929","citing_title":"HippoSpark: An On-Demand Experience System for LLM Reasoning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15384","citing_title":"Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2507.13334","citing_title":"A Survey of Context Engineering for Large Language Models","ref_index":281,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB","json":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB.json","graph_json":"https://pith.science/api/pith-number/WJFUAE6BJU3BEVAVFZHKE5N5YB/graph.json","events_json":"https://pith.science/api/pith-number/WJFUAE6BJU3BEVAVFZHKE5N5YB/events.json","paper":"https://pith.science/paper/WJFUAE6B"},"agent_actions":{"view_html":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB","download_json":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB.json","view_paper":"https://pith.science/paper/WJFUAE6B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.17716&json=true","fetch_graph":"https://pith.science/api/pith-number/WJFUAE6BJU3BEVAVFZHKE5N5YB/graph.json","fetch_events":"https://pith.science/api/pith-number/WJFUAE6BJU3BEVAVFZHKE5N5YB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB/action/storage_attestation","attest_author":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB/action/author_attestation","sign_citation":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB/action/citation_signature","submit_replication":"https://pith.science/pith/WJFUAE6BJU3BEVAVFZHKE5N5YB/action/replication_record"}},"created_at":"2026-07-05T11:08:30.781608+00:00","updated_at":"2026-07-05T11:08:30.781608+00:00"}