{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:O6FWM2P6FE7DJW5GD4RMSKXQP5","short_pith_number":"pith:O6FWM2P6","canonical_record":{"source":{"id":"2404.17833","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-04-27T08:56:45Z","cross_cats_sorted":["cs.PL"],"title_canon_sha256":"27732ecbd965165962f6b57489cdae3311289956ea2ac0de3549731f9d529968","abstract_canon_sha256":"4d3019a65e4282e6267a056a21e0a6132efefea937a14c8f6e5c77936032ff17"},"schema_version":"1.0"},"canonical_sha256":"778b6669fe293e34dba61f22c92af07f5c3970f7246a062e5aa1ab527f2ca25e","source":{"kind":"arxiv","id":"2404.17833","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.17833","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"arxiv_version","alias_value":"2404.17833v1","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.17833","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_12","alias_value":"O6FWM2P6FE7D","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_16","alias_value":"O6FWM2P6FE7DJW5G","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_8","alias_value":"O6FWM2P6","created_at":"2026-07-05T08:12:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:O6FWM2P6FE7DJW5GD4RMSKXQP5","target":"record","payload":{"canonical_record":{"source":{"id":"2404.17833","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-04-27T08:56:45Z","cross_cats_sorted":["cs.PL"],"title_canon_sha256":"27732ecbd965165962f6b57489cdae3311289956ea2ac0de3549731f9d529968","abstract_canon_sha256":"4d3019a65e4282e6267a056a21e0a6132efefea937a14c8f6e5c77936032ff17"},"schema_version":"1.0"},"canonical_sha256":"778b6669fe293e34dba61f22c92af07f5c3970f7246a062e5aa1ab527f2ca25e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:54.752941Z","signature_b64":"KrZqm3UCSZXbgwiSyHKPuMZgWqeQbSdQHWgzF/b9Vvf/C9XWLnkn6NWsoofn8nUePtQpU8LDSHyvOOn4JBmsCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"778b6669fe293e34dba61f22c92af07f5c3970f7246a062e5aa1ab527f2ca25e","last_reissued_at":"2026-07-05T08:12:54.752408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:54.752408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.17833","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:12:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J1CjEzzbCOTglwHUZK/tD6KEVpSXhC++muKDxuu5JG2NUd1JEn+ISGGbrjGpr1OTt0CrnzxG8CdekYfxU3U9Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:09:41.250581Z"},"content_sha256":"638ca757eabffb8c5dcd227c4470d35a536d6484603a837afcd79b20b688865e","schema_version":"1.0","event_id":"sha256:638ca757eabffb8c5dcd227c4470d35a536d6484603a837afcd79b20b688865e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:O6FWM2P6FE7DJW5GD4RMSKXQP5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.PL"],"primary_cat":"cs.AI","authors_text":"Daoyuan Wu, Pingchuan Ma, Shuai Wang, Zhenlan Ji, Zongjie Li","submitted_at":"2024-04-27T08:56:45Z","abstract_excerpt":"Agents based on large language models (LLMs) have demonstrated effectiveness in solving a wide range of tasks by integrating LLMs with key modules such as planning, memory, and tool usage. Increasingly, customers are adopting LLM agents across a variety of commercial applications critical to reliability, including support for mental well-being, chemical synthesis, and software development. Nevertheless, our observations and daily use of LLM agents indicate that they are prone to making erroneous plans, especially when the tasks are complex and require long-term planning.\n  In this paper, we pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.17833","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/2404.17833/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:12:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9JrLRAnAcBuK2+sD7OKCiD38u+N+eGKPG08AK5s3fXdd6cXdM6DKnib7eLeqKPARG8CqZrlPVpVdfDVdi2cOBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:09:41.251130Z"},"content_sha256":"f5a1b452e3a83cd4232d5500c9bec2961a2183dd23fed6d2af930de842adadfe","schema_version":"1.0","event_id":"sha256:f5a1b452e3a83cd4232d5500c9bec2961a2183dd23fed6d2af930de842adadfe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/bundle.json","state_url":"https://pith.science/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T22:09:41Z","links":{"resolver":"https://pith.science/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5","bundle":"https://pith.science/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/bundle.json","state":"https://pith.science/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/O6FWM2P6FE7DJW5GD4RMSKXQP5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:O6FWM2P6FE7DJW5GD4RMSKXQP5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4d3019a65e4282e6267a056a21e0a6132efefea937a14c8f6e5c77936032ff17","cross_cats_sorted":["cs.PL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-04-27T08:56:45Z","title_canon_sha256":"27732ecbd965165962f6b57489cdae3311289956ea2ac0de3549731f9d529968"},"schema_version":"1.0","source":{"id":"2404.17833","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.17833","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"arxiv_version","alias_value":"2404.17833v1","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.17833","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_12","alias_value":"O6FWM2P6FE7D","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_16","alias_value":"O6FWM2P6FE7DJW5G","created_at":"2026-07-05T08:12:54Z"},{"alias_kind":"pith_short_8","alias_value":"O6FWM2P6","created_at":"2026-07-05T08:12:54Z"}],"graph_snapshots":[{"event_id":"sha256:f5a1b452e3a83cd4232d5500c9bec2961a2183dd23fed6d2af930de842adadfe","target":"graph","created_at":"2026-07-05T08:12:54Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2404.17833/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Agents based on large language models (LLMs) have demonstrated effectiveness in solving a wide range of tasks by integrating LLMs with key modules such as planning, memory, and tool usage. Increasingly, customers are adopting LLM agents across a variety of commercial applications critical to reliability, including support for mental well-being, chemical synthesis, and software development. Nevertheless, our observations and daily use of LLM agents indicate that they are prone to making erroneous plans, especially when the tasks are complex and require long-term planning.\n  In this paper, we pr","authors_text":"Daoyuan Wu, Pingchuan Ma, Shuai Wang, Zhenlan Ji, Zongjie Li","cross_cats":["cs.PL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-04-27T08:56:45Z","title":"Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.17833","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:638ca757eabffb8c5dcd227c4470d35a536d6484603a837afcd79b20b688865e","target":"record","created_at":"2026-07-05T08:12:54Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"4d3019a65e4282e6267a056a21e0a6132efefea937a14c8f6e5c77936032ff17","cross_cats_sorted":["cs.PL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-04-27T08:56:45Z","title_canon_sha256":"27732ecbd965165962f6b57489cdae3311289956ea2ac0de3549731f9d529968"},"schema_version":"1.0","source":{"id":"2404.17833","kind":"arxiv","version":1}},"canonical_sha256":"778b6669fe293e34dba61f22c92af07f5c3970f7246a062e5aa1ab527f2ca25e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"778b6669fe293e34dba61f22c92af07f5c3970f7246a062e5aa1ab527f2ca25e","first_computed_at":"2026-07-05T08:12:54.752408Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:12:54.752408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KrZqm3UCSZXbgwiSyHKPuMZgWqeQbSdQHWgzF/b9Vvf/C9XWLnkn6NWsoofn8nUePtQpU8LDSHyvOOn4JBmsCA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:12:54.752941Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.17833","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:638ca757eabffb8c5dcd227c4470d35a536d6484603a837afcd79b20b688865e","sha256:f5a1b452e3a83cd4232d5500c9bec2961a2183dd23fed6d2af930de842adadfe"],"state_sha256":"050ab3420a1709e4472d86734f9aa0ecbfd15a2eb446214c6077f6c2bb856e2f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f7hXtyPlBqu2DyfjXXCoBm1Ymyr79cLv6vebaLF0r1CEYKz2LbgrrWzonYGrXPifIRZ0XGsEnjwCjuFQnZw8Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:09:41.255057Z","bundle_sha256":"3354176b68fbef660153263d42af3a050e6dd41ed951b9e0f86a90ac070bcf96"}}