{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CCV7M372PYXA3ZWSIYHMOFRZLJ","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":"f37b8920df21ddeb3450383ccdb71e796593be21082a3db14afd43f17003ab04","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-30T15:58:41Z","title_canon_sha256":"54d1d69943d3ca3b218b17bb85c03c9dff97298483bd258964089b617c95ddf5"},"schema_version":"1.0","source":{"id":"2412.21033","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.21033","created_at":"2026-07-05T11:37:14Z"},{"alias_kind":"arxiv_version","alias_value":"2412.21033v2","created_at":"2026-07-05T11:37:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.21033","created_at":"2026-07-05T11:37:14Z"},{"alias_kind":"pith_short_12","alias_value":"CCV7M372PYXA","created_at":"2026-07-05T11:37:14Z"},{"alias_kind":"pith_short_16","alias_value":"CCV7M372PYXA3ZWS","created_at":"2026-07-05T11:37:14Z"},{"alias_kind":"pith_short_8","alias_value":"CCV7M372","created_at":"2026-07-05T11:37:14Z"}],"graph_snapshots":[{"event_id":"sha256:a1f49355bd09ef18c1461ab5a58fd329b75c4b53b4d5a4fbb966ea4429c070a3","target":"graph","created_at":"2026-07-05T11:37:14Z","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/2412.21033/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present Plancraft, a multi-modal evaluation dataset for LLM agents. Plancraft has both a text-only and multi-modal interface, based on the Minecraft crafting GUI. We include the Minecraft Wiki to evaluate tool use and Retrieval Augmented Generation (RAG), as well as a handcrafted planner and Oracle Retriever, to ablate the different components of a modern agent architecture. To evaluate decision-making, Plancraft also includes a subset of examples that are intentionally unsolvable, providing a realistic challenge that requires the agent not only to complete tasks but also to decide whether ","authors_text":"Alex Lascarides, Frank Keller, Gautier Dagan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-30T15:58:41Z","title":"Plancraft: an evaluation dataset for planning with LLM agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.21033","kind":"arxiv","version":2},"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":"sha256:4ffded63c78e3c266a8855ce59b1a5b5c14fc6417cfbe0816006a3c87789c785","target":"record","created_at":"2026-07-05T11:37:14Z","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":"f37b8920df21ddeb3450383ccdb71e796593be21082a3db14afd43f17003ab04","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-30T15:58:41Z","title_canon_sha256":"54d1d69943d3ca3b218b17bb85c03c9dff97298483bd258964089b617c95ddf5"},"schema_version":"1.0","source":{"id":"2412.21033","kind":"arxiv","version":2}},"canonical_sha256":"10abf66ffa7e2e0de6d2460ec716395a7b406fb2696ce219dfb33152b29a6a68","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"10abf66ffa7e2e0de6d2460ec716395a7b406fb2696ce219dfb33152b29a6a68","first_computed_at":"2026-07-05T11:37:14.449759Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:37:14.449759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VUlq2CV4o7wuS+TmxYZST8jXdkTDbOJGypJNM1quNyh0PM6n1q0Yr88pL7M+uSGreLByZm9ZIoRGnzBeEmQQBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:37:14.450246Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.21033","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ffded63c78e3c266a8855ce59b1a5b5c14fc6417cfbe0816006a3c87789c785","sha256:a1f49355bd09ef18c1461ab5a58fd329b75c4b53b4d5a4fbb966ea4429c070a3"],"state_sha256":"da2537bb32fce89796f594c7ed2da72f70c03f92641793f3823be48035712ac5"}