{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:CC5SUUNJUADNREYZ25NXPXNFXG","short_pith_number":"pith:CC5SUUNJ","canonical_record":{"source":{"id":"2605.27957","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T04:50:23Z","cross_cats_sorted":[],"title_canon_sha256":"3ae6d032a730d44400e3f58fa8fc4e676e1324e4a64134361eb08abed18cec48","abstract_canon_sha256":"9db734ebe98b9b7e6c34f94d5ae2659dcde45cabf693e68b84722f25e0d24747"},"schema_version":"1.0"},"canonical_sha256":"10bb2a51a9a006d89319d75b77dda5b9ad91c58e1670bbfcaf9e54db8490c24e","source":{"kind":"arxiv","id":"2605.27957","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27957","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27957v1","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27957","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_12","alias_value":"CC5SUUNJUADN","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_16","alias_value":"CC5SUUNJUADNREYZ","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_8","alias_value":"CC5SUUNJ","created_at":"2026-05-28T01:04:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:CC5SUUNJUADNREYZ25NXPXNFXG","target":"record","payload":{"canonical_record":{"source":{"id":"2605.27957","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T04:50:23Z","cross_cats_sorted":[],"title_canon_sha256":"3ae6d032a730d44400e3f58fa8fc4e676e1324e4a64134361eb08abed18cec48","abstract_canon_sha256":"9db734ebe98b9b7e6c34f94d5ae2659dcde45cabf693e68b84722f25e0d24747"},"schema_version":"1.0"},"canonical_sha256":"10bb2a51a9a006d89319d75b77dda5b9ad91c58e1670bbfcaf9e54db8490c24e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:54.147796Z","signature_b64":"Sbb2dINpx0XV7q8/fhqY8aB34gkjT97XIiRoCgz7rwtSP75DPTquV4xssagIiNFHgOS0Px10RPyZyhMXJ77aCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10bb2a51a9a006d89319d75b77dda5b9ad91c58e1670bbfcaf9e54db8490c24e","last_reissued_at":"2026-05-28T01:04:54.147381Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:54.147381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.27957","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":"2026-05-28T01:04:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eOaQ+xtAiQse+jiasvxUWQZMWndgTMXYFUs/W+8IkxMppDRU2znfRZqNdeudfGfC666xgMx8aQ5w9v43/zrdBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T03:43:56.597260Z"},"content_sha256":"52f3bcacd1a27327290bb5a458fcb5c417fcf72e6fef300e13677edf911a801d","schema_version":"1.0","event_id":"sha256:52f3bcacd1a27327290bb5a458fcb5c417fcf72e6fef300e13677edf911a801d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:CC5SUUNJUADNREYZ25NXPXNFXG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DisasterBench: Benchmarking LLM Planning under Typed Tool Interface Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Mostafavi, Chengkai Liu, James Caverlee, Kai Yin, Weifeng Zhang, Xiangjue Dong, Yiming Xiao, Zhewei Liu, Zhitong Chen, Zhiyuan Wang","submitted_at":"2026-05-27T04:50:23Z","abstract_excerpt":"Disasters cause severe societal impacts, demanding rapid coordination of heterogeneous AI tools, from satellite analysis to flood prediction and damage assessment, into coherent multi-step workflows. As LLMs increasingly serve as orchestrators of such pipelines, effective coordination requires more than selecting semantically plausible tools: LLMs must generate executable workflows with correct parameter binding and dependency propagation.\n  We introduce DisasterBench, a benchmark for evaluating structured multi-agent planning over semantically similar but operationally distinct disaster-respo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27957","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/2605.27957/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":"2026-05-28T01:04:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3wFEugCDRUWojGsTwe64LQVjlfb876P2jY1fEQ9q/VUZNk6PlRr3MgS57J+Q0b5qVYwiTt3A5sUNl8MAzx4GCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T03:43:56.598041Z"},"content_sha256":"2c010113f06c4fcd97665c20027f9d940f783818e12ee12ffc6f08cf3137bcf1","schema_version":"1.0","event_id":"sha256:2c010113f06c4fcd97665c20027f9d940f783818e12ee12ffc6f08cf3137bcf1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CC5SUUNJUADNREYZ25NXPXNFXG/bundle.json","state_url":"https://pith.science/pith/CC5SUUNJUADNREYZ25NXPXNFXG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CC5SUUNJUADNREYZ25NXPXNFXG/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":"2026-06-07T03:43:56Z","links":{"resolver":"https://pith.science/pith/CC5SUUNJUADNREYZ25NXPXNFXG","bundle":"https://pith.science/pith/CC5SUUNJUADNREYZ25NXPXNFXG/bundle.json","state":"https://pith.science/pith/CC5SUUNJUADNREYZ25NXPXNFXG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CC5SUUNJUADNREYZ25NXPXNFXG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CC5SUUNJUADNREYZ25NXPXNFXG","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":"9db734ebe98b9b7e6c34f94d5ae2659dcde45cabf693e68b84722f25e0d24747","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T04:50:23Z","title_canon_sha256":"3ae6d032a730d44400e3f58fa8fc4e676e1324e4a64134361eb08abed18cec48"},"schema_version":"1.0","source":{"id":"2605.27957","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27957","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27957v1","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27957","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_12","alias_value":"CC5SUUNJUADN","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_16","alias_value":"CC5SUUNJUADNREYZ","created_at":"2026-05-28T01:04:54Z"},{"alias_kind":"pith_short_8","alias_value":"CC5SUUNJ","created_at":"2026-05-28T01:04:54Z"}],"graph_snapshots":[{"event_id":"sha256:2c010113f06c4fcd97665c20027f9d940f783818e12ee12ffc6f08cf3137bcf1","target":"graph","created_at":"2026-05-28T01:04:54Z","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/2605.27957/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Disasters cause severe societal impacts, demanding rapid coordination of heterogeneous AI tools, from satellite analysis to flood prediction and damage assessment, into coherent multi-step workflows. As LLMs increasingly serve as orchestrators of such pipelines, effective coordination requires more than selecting semantically plausible tools: LLMs must generate executable workflows with correct parameter binding and dependency propagation.\n  We introduce DisasterBench, a benchmark for evaluating structured multi-agent planning over semantically similar but operationally distinct disaster-respo","authors_text":"Ali Mostafavi, Chengkai Liu, James Caverlee, Kai Yin, Weifeng Zhang, Xiangjue Dong, Yiming Xiao, Zhewei Liu, Zhitong Chen, Zhiyuan Wang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T04:50:23Z","title":"DisasterBench: Benchmarking LLM Planning under Typed Tool Interface Constraints"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27957","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":"sha256:52f3bcacd1a27327290bb5a458fcb5c417fcf72e6fef300e13677edf911a801d","target":"record","created_at":"2026-05-28T01:04:54Z","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":"9db734ebe98b9b7e6c34f94d5ae2659dcde45cabf693e68b84722f25e0d24747","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T04:50:23Z","title_canon_sha256":"3ae6d032a730d44400e3f58fa8fc4e676e1324e4a64134361eb08abed18cec48"},"schema_version":"1.0","source":{"id":"2605.27957","kind":"arxiv","version":1}},"canonical_sha256":"10bb2a51a9a006d89319d75b77dda5b9ad91c58e1670bbfcaf9e54db8490c24e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"10bb2a51a9a006d89319d75b77dda5b9ad91c58e1670bbfcaf9e54db8490c24e","first_computed_at":"2026-05-28T01:04:54.147381Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:54.147381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Sbb2dINpx0XV7q8/fhqY8aB34gkjT97XIiRoCgz7rwtSP75DPTquV4xssagIiNFHgOS0Px10RPyZyhMXJ77aCQ==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:54.147796Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.27957","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52f3bcacd1a27327290bb5a458fcb5c417fcf72e6fef300e13677edf911a801d","sha256:2c010113f06c4fcd97665c20027f9d940f783818e12ee12ffc6f08cf3137bcf1"],"state_sha256":"d20c8491cdeaae3ba4ab3148f0e2e28f4dcced2a86631dde5333fa80ef36ae40"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ew44B3BflcB0f/PvrIYeOrZZVGPhmNssYQ40/XV58btu1y//amZTansc8OVlcJFYF1Xe1QJBBEb9V09wuH3jCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T03:43:56.601614Z","bundle_sha256":"913e387306943fe2ae101f84443b718e2650e19f95ad8a6b5e6a70d92794ef66"}}