{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3KCOGAPKDJUQP2OE22JKH5KZDB","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":"5e7354d42ffa530a8b2253f9842085c067378e4b8970af5d70a70caea254b80e","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-27T07:13:29Z","title_canon_sha256":"02e6bd34daa12dcc23b8ab7c96b97fbd4e8b11d96438d61fec7a3e1b2bae6dec"},"schema_version":"1.0","source":{"id":"2506.21967","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.21967","created_at":"2026-07-05T11:28:19Z"},{"alias_kind":"arxiv_version","alias_value":"2506.21967v1","created_at":"2026-07-05T11:28:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.21967","created_at":"2026-07-05T11:28:19Z"},{"alias_kind":"pith_short_12","alias_value":"3KCOGAPKDJUQ","created_at":"2026-07-05T11:28:19Z"},{"alias_kind":"pith_short_16","alias_value":"3KCOGAPKDJUQP2OE","created_at":"2026-07-05T11:28:19Z"},{"alias_kind":"pith_short_8","alias_value":"3KCOGAPK","created_at":"2026-07-05T11:28:19Z"}],"graph_snapshots":[{"event_id":"sha256:5e017d3ebba64eed26b2a3b1a5a67c14734fa5ead12381e28dc6c96ab910754e","target":"graph","created_at":"2026-07-05T11:28:19Z","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/2506.21967/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current evaluations of tool-integrated LLM agents typically focus on end-to-end tool-usage evaluation while neglecting their stability. This limits their real-world applicability, as various internal or external factors can cause agents to crash or behave abnormally. Our research addresses this by investigating whether agents are vulnerable to errors throughout the entire tool invocation process, including reading tool documentation, selecting tools and generating parameters, and processing the tool's response. Through extensive experiments, we observe that agents are highly susceptible to err","authors_text":"Hanchao Liu, Ke Wang, Sai Zhou, Sujian Li, Weimin Xiong, Wei Peng, Yifan Song","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-27T07:13:29Z","title":"More Vulnerable than You Think: On the Stability of Tool-Integrated LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.21967","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":"sha256:2b5a2223c103484ed62ded0b2eaea87395ad6ea2b871857eba877b3f9a66863b","target":"record","created_at":"2026-07-05T11:28:19Z","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":"5e7354d42ffa530a8b2253f9842085c067378e4b8970af5d70a70caea254b80e","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-27T07:13:29Z","title_canon_sha256":"02e6bd34daa12dcc23b8ab7c96b97fbd4e8b11d96438d61fec7a3e1b2bae6dec"},"schema_version":"1.0","source":{"id":"2506.21967","kind":"arxiv","version":1}},"canonical_sha256":"da84e301ea1a6907e9c4d692a3f559185c908a522b48e3095e7245b8a085aee5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da84e301ea1a6907e9c4d692a3f559185c908a522b48e3095e7245b8a085aee5","first_computed_at":"2026-07-05T11:28:19.393670Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:28:19.393670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"l8jEP33J7zyTkzPsRj8g0iAc9zZevRBe1F9/5vhQIt3YE+lChWmTSt7MoOivAAHaAtB/wxDpXfjfQoX/gXjqAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:28:19.394128Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.21967","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2b5a2223c103484ed62ded0b2eaea87395ad6ea2b871857eba877b3f9a66863b","sha256:5e017d3ebba64eed26b2a3b1a5a67c14734fa5ead12381e28dc6c96ab910754e"],"state_sha256":"aca6bb9b485e3a8b23769314ff5738f9ceda91ab9551ae388926044845bd6220"}