{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ASCGQLUJFA6PLLYOGBIKQKYMVZ","short_pith_number":"pith:ASCGQLUJ","schema_version":"1.0","canonical_sha256":"0484682e89283cf5af0e3050a82b0cae47ecd1c2efba6e29396f63a770762457","source":{"kind":"arxiv","id":"2505.13546","version":1},"attestation_state":"computed","paper":{"title":"Prompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.MA"],"primary_cat":"cs.AI","authors_text":"Haohan Wang, Ke Chen, Xitong Zhang, Yufei Zhou","submitted_at":"2025-05-19T03:28:33Z","abstract_excerpt":"Automatic prompt generation plays a crucial role in enabling general-purpose multi-agent systems to perform diverse tasks autonomously. Existing methods typically evaluate prompts based on their immediate task performance, overlooking the intrinsic qualities that determine their reliability. This outcome-centric view not only limits interpretability but also fails to account for the inherent stochasticity of large language models (LLMs). In this work, we bring attention to prompt stability-the consistency of model responses across repeated executions-as a key factor for building robust and eff"},"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":"2505.13546","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-19T03:28:33Z","cross_cats_sorted":["cs.CL","cs.MA"],"title_canon_sha256":"cb7c51d502c06e7ff28d69310c80e6b009229121f201000a80c54858084c7fc9","abstract_canon_sha256":"03b50fdb87502d1c79ab17cfbd332a010428d9092785e5607cbe90700757e9fe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:13.882846Z","signature_b64":"4ABm6ZYUSZEJ4jxNGimE13lpeaYYVRGo0jqe7cxaGVRilXPBhNVai/Rbb22fzkif1+kXJikyuNGDsPBolt8ADQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0484682e89283cf5af0e3050a82b0cae47ecd1c2efba6e29396f63a770762457","last_reissued_at":"2026-07-05T11:06:13.882099Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:13.882099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.MA"],"primary_cat":"cs.AI","authors_text":"Haohan Wang, Ke Chen, Xitong Zhang, Yufei Zhou","submitted_at":"2025-05-19T03:28:33Z","abstract_excerpt":"Automatic prompt generation plays a crucial role in enabling general-purpose multi-agent systems to perform diverse tasks autonomously. Existing methods typically evaluate prompts based on their immediate task performance, overlooking the intrinsic qualities that determine their reliability. This outcome-centric view not only limits interpretability but also fails to account for the inherent stochasticity of large language models (LLMs). In this work, we bring attention to prompt stability-the consistency of model responses across repeated executions-as a key factor for building robust and eff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13546","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/2505.13546/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":"2505.13546","created_at":"2026-07-05T11:06:13.882201+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.13546v1","created_at":"2026-07-05T11:06:13.882201+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13546","created_at":"2026-07-05T11:06:13.882201+00:00"},{"alias_kind":"pith_short_12","alias_value":"ASCGQLUJFA6P","created_at":"2026-07-05T11:06:13.882201+00:00"},{"alias_kind":"pith_short_16","alias_value":"ASCGQLUJFA6PLLYO","created_at":"2026-07-05T11:06:13.882201+00:00"},{"alias_kind":"pith_short_8","alias_value":"ASCGQLUJ","created_at":"2026-07-05T11:06:13.882201+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.21035","citing_title":"GenoMAS: A Multi-Agent Framework for Scientific Discovery via Code-Driven Gene Expression Analysis","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2603.10477","citing_title":"PEEM: Prompt Engineering Evaluation Metrics for Interpretable Joint Evaluation of Prompts and Responses","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ","json":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ.json","graph_json":"https://pith.science/api/pith-number/ASCGQLUJFA6PLLYOGBIKQKYMVZ/graph.json","events_json":"https://pith.science/api/pith-number/ASCGQLUJFA6PLLYOGBIKQKYMVZ/events.json","paper":"https://pith.science/paper/ASCGQLUJ"},"agent_actions":{"view_html":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ","download_json":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ.json","view_paper":"https://pith.science/paper/ASCGQLUJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.13546&json=true","fetch_graph":"https://pith.science/api/pith-number/ASCGQLUJFA6PLLYOGBIKQKYMVZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ASCGQLUJFA6PLLYOGBIKQKYMVZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ/action/storage_attestation","attest_author":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ/action/author_attestation","sign_citation":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ/action/citation_signature","submit_replication":"https://pith.science/pith/ASCGQLUJFA6PLLYOGBIKQKYMVZ/action/replication_record"}},"created_at":"2026-07-05T11:06:13.882201+00:00","updated_at":"2026-07-05T11:06:13.882201+00:00"}