{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:N64CVEHQLLCZ3BR6JRUT2Z3J3J","short_pith_number":"pith:N64CVEHQ","canonical_record":{"source":{"id":"2504.14773","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T00:02:50Z","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"title_canon_sha256":"22e92021edfeb22170bb96e7577b2c97d5ca5ebedbea7becac053a1d26342092","abstract_canon_sha256":"94bff3baae32681c4cb8bb4972e5d10b81c4e177f15f416cb08eb0a57de9b37f"},"schema_version":"1.0"},"canonical_sha256":"6fb82a90f05ac59d863e4c693d6769da4cde417fb2813ecd55cf63800ea736f0","source":{"kind":"arxiv","id":"2504.14773","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.14773","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"arxiv_version","alias_value":"2504.14773v1","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.14773","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_12","alias_value":"N64CVEHQLLCZ","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_16","alias_value":"N64CVEHQLLCZ3BR6","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_8","alias_value":"N64CVEHQ","created_at":"2026-07-05T10:51:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:N64CVEHQLLCZ3BR6JRUT2Z3J3J","target":"record","payload":{"canonical_record":{"source":{"id":"2504.14773","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T00:02:50Z","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"title_canon_sha256":"22e92021edfeb22170bb96e7577b2c97d5ca5ebedbea7becac053a1d26342092","abstract_canon_sha256":"94bff3baae32681c4cb8bb4972e5d10b81c4e177f15f416cb08eb0a57de9b37f"},"schema_version":"1.0"},"canonical_sha256":"6fb82a90f05ac59d863e4c693d6769da4cde417fb2813ecd55cf63800ea736f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:50.387014Z","signature_b64":"4Sr7mKaVamoliwYn7bqHq5BBdGyVXWRX9kuWMNJK4tkCF5Q92JWJrx8lHzvEV2g77FPc+e78DlqY4XlIG7MlCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6fb82a90f05ac59d863e4c693d6769da4cde417fb2813ecd55cf63800ea736f0","last_reissued_at":"2026-07-05T10:51:50.386554Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:50.386554Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.14773","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":"2026-07-05T10:51:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TWWrsie8wfNeD3yD0Ld+NQXmr3TLqVn/b+IaOCys3tiiPnFtr9oZ7c7BpTVxhNzKXEMeEfPEXu7W90Vog15zCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:30:07.655542Z"},"content_sha256":"2d2bed2abb8cc32d029b6ea90f2ca15e432c02fe58ac556eaa3acfc10eff2eb3","schema_version":"1.0","event_id":"sha256:2d2bed2abb8cc32d029b6ea90f2ca15e432c02fe58ac556eaa3acfc10eff2eb3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:N64CVEHQLLCZ3BR6JRUT2Z3J3J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PLANET: A Collection of Benchmarks for Evaluating LLMs' Planning Capabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Fei Liu, Haoming Li, Jonathan Zhang, Zhaoliang Chen","submitted_at":"2025-04-21T00:02:50Z","abstract_excerpt":"Planning is central to agents and agentic AI. The ability to plan, e.g., creating travel itineraries within a budget, holds immense potential in both scientific and commercial contexts. Moreover, optimal plans tend to require fewer resources compared to ad-hoc methods. To date, a comprehensive understanding of existing planning benchmarks appears to be lacking. Without it, comparing planning algorithms' performance across domains or selecting suitable algorithms for new scenarios remains challenging. In this paper, we examine a range of planning benchmarks to identify commonly used testbeds fo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.14773","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/2504.14773/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":"2026-07-05T10:51:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aWyAPo1Q5Q1eFYmWapEU+NpeX3SroQ+VqTnPrDwQ51+WQsbjzhJ8ctRQee0wz8UC/cvA1icpCXO1giJSHw/bDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:30:07.662473Z"},"content_sha256":"654b8984ac8089db71fc8cf5538d42c21d93bedf6119d2906bfd1b7d3e19fd32","schema_version":"1.0","event_id":"sha256:654b8984ac8089db71fc8cf5538d42c21d93bedf6119d2906bfd1b7d3e19fd32"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/bundle.json","state_url":"https://pith.science/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/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":"2026-08-17T12:30:07Z","links":{"resolver":"https://pith.science/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J","bundle":"https://pith.science/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/bundle.json","state":"https://pith.science/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N64CVEHQLLCZ3BR6JRUT2Z3J3J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:N64CVEHQLLCZ3BR6JRUT2Z3J3J","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":"94bff3baae32681c4cb8bb4972e5d10b81c4e177f15f416cb08eb0a57de9b37f","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T00:02:50Z","title_canon_sha256":"22e92021edfeb22170bb96e7577b2c97d5ca5ebedbea7becac053a1d26342092"},"schema_version":"1.0","source":{"id":"2504.14773","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.14773","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"arxiv_version","alias_value":"2504.14773v1","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.14773","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_12","alias_value":"N64CVEHQLLCZ","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_16","alias_value":"N64CVEHQLLCZ3BR6","created_at":"2026-07-05T10:51:50Z"},{"alias_kind":"pith_short_8","alias_value":"N64CVEHQ","created_at":"2026-07-05T10:51:50Z"}],"graph_snapshots":[{"event_id":"sha256:654b8984ac8089db71fc8cf5538d42c21d93bedf6119d2906bfd1b7d3e19fd32","target":"graph","created_at":"2026-07-05T10:51:50Z","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/2504.14773/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Planning is central to agents and agentic AI. The ability to plan, e.g., creating travel itineraries within a budget, holds immense potential in both scientific and commercial contexts. Moreover, optimal plans tend to require fewer resources compared to ad-hoc methods. To date, a comprehensive understanding of existing planning benchmarks appears to be lacking. Without it, comparing planning algorithms' performance across domains or selecting suitable algorithms for new scenarios remains challenging. In this paper, we examine a range of planning benchmarks to identify commonly used testbeds fo","authors_text":"Fei Liu, Haoming Li, Jonathan Zhang, Zhaoliang Chen","cross_cats":["cs.CL","cs.LG","cs.MA"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T00:02:50Z","title":"PLANET: A Collection of Benchmarks for Evaluating LLMs' Planning Capabilities"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.14773","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":"sha256:2d2bed2abb8cc32d029b6ea90f2ca15e432c02fe58ac556eaa3acfc10eff2eb3","target":"record","created_at":"2026-07-05T10:51:50Z","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":"94bff3baae32681c4cb8bb4972e5d10b81c4e177f15f416cb08eb0a57de9b37f","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T00:02:50Z","title_canon_sha256":"22e92021edfeb22170bb96e7577b2c97d5ca5ebedbea7becac053a1d26342092"},"schema_version":"1.0","source":{"id":"2504.14773","kind":"arxiv","version":1}},"canonical_sha256":"6fb82a90f05ac59d863e4c693d6769da4cde417fb2813ecd55cf63800ea736f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6fb82a90f05ac59d863e4c693d6769da4cde417fb2813ecd55cf63800ea736f0","first_computed_at":"2026-07-05T10:51:50.386554Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:51:50.386554Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4Sr7mKaVamoliwYn7bqHq5BBdGyVXWRX9kuWMNJK4tkCF5Q92JWJrx8lHzvEV2g77FPc+e78DlqY4XlIG7MlCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:51:50.387014Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.14773","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2d2bed2abb8cc32d029b6ea90f2ca15e432c02fe58ac556eaa3acfc10eff2eb3","sha256:654b8984ac8089db71fc8cf5538d42c21d93bedf6119d2906bfd1b7d3e19fd32"],"state_sha256":"4225a772d5101974a4ec0dbf87b0c16201b5344c2279bf42497de0ae175110f6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UXAyUaOgbMYTXJk0QnbsIbqbga1erSlhnS007qyW+0SnKl/UvEw2NPbNY+JyeTyvKzJVtaUCzSJCS13V9LHMDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T12:30:07.682878Z","bundle_sha256":"71a0763340143aef66daeb6b163e0c3a524173722b80444483d4a6d88465a754"}}