{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VBKDFNJSLLFIIBXZRQD7ATCCL6","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":"8d42f50540386fde7dfbfca7204631b6608ff8d1c4ee4e0f7f9659fffe247f79","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T05:31:50Z","title_canon_sha256":"814b298b05f5c0b6255f223379792f4aedc6d39818cd548bbe415f7c8212fa81"},"schema_version":"1.0","source":{"id":"2506.04649","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04649","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04649v1","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04649","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_12","alias_value":"VBKDFNJSLLFI","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_16","alias_value":"VBKDFNJSLLFIIBXZ","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_8","alias_value":"VBKDFNJS","created_at":"2026-07-05T11:16:18Z"}],"graph_snapshots":[{"event_id":"sha256:44fff961cd70ce3cc10655286462a5927f24845f028f60f2128dcff243d38fb3","target":"graph","created_at":"2026-07-05T11:16:18Z","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/2506.04649/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Real-world planning problems require constant adaptation to changing requirements and balancing of competing constraints. However, current benchmarks for evaluating LLMs' planning capabilities primarily focus on static, single-turn scenarios. We introduce Flex-TravelPlanner, a benchmark that evaluates language models' ability to reason flexibly in dynamic planning scenarios. Building on the TravelPlanner dataset~\\citep{xie2024travelplanner}, we introduce two novel evaluation settings: (1) sequential constraint introduction across multiple turns, and (2) scenarios with explicitly prioritized co","authors_text":"Alice Oh, Eunsu Kim, Juhyun Oh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T05:31:50Z","title":"Flex-TravelPlanner: A Benchmark for Flexible Planning with Language Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04649","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":"sha256:4c3fd9e14a56054b0c0ad12040aa29cc5bf63abd4804a626a916880bc0c0a9a7","target":"record","created_at":"2026-07-05T11:16:18Z","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":"8d42f50540386fde7dfbfca7204631b6608ff8d1c4ee4e0f7f9659fffe247f79","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T05:31:50Z","title_canon_sha256":"814b298b05f5c0b6255f223379792f4aedc6d39818cd548bbe415f7c8212fa81"},"schema_version":"1.0","source":{"id":"2506.04649","kind":"arxiv","version":1}},"canonical_sha256":"a85432b5325aca8406f98c07f04c425f8854650346cc088a609258542bce5260","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a85432b5325aca8406f98c07f04c425f8854650346cc088a609258542bce5260","first_computed_at":"2026-07-05T11:16:18.458719Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:18.458719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"unRest2+UTuBWRYBYL/9T+32vGYWpDZ6zklDyh5NeLlEelTW5jZX9+HZWgdoV7qVgRHEEVATycJmDHfbR+ufDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:18.459133Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04649","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4c3fd9e14a56054b0c0ad12040aa29cc5bf63abd4804a626a916880bc0c0a9a7","sha256:44fff961cd70ce3cc10655286462a5927f24845f028f60f2128dcff243d38fb3"],"state_sha256":"5265a54fbe646aed0d5a0caa054831f95cb2b98786cf60ea130a8cf50226cf19"}