{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:F5LM7OP2EO26MTKT34ITXHY35P","short_pith_number":"pith:F5LM7OP2","canonical_record":{"source":{"id":"2410.03907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T20:21:40Z","cross_cats_sorted":[],"title_canon_sha256":"94462c6328a1b78451e32a6ed8e09e99d2e2e0f0091c4b4d6867df568092a2b5","abstract_canon_sha256":"263ccda2da4aaf44d2156c6249561831c8a535ab4950f17237545ccf121dcbe8"},"schema_version":"1.0"},"canonical_sha256":"2f56cfb9fa23b5e64d53df113b9f1bebe21a49049a2debc1f304787a523a4852","source":{"kind":"arxiv","id":"2410.03907","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03907","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03907v1","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03907","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_12","alias_value":"F5LM7OP2EO26","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_16","alias_value":"F5LM7OP2EO26MTKT","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_8","alias_value":"F5LM7OP2","created_at":"2026-07-05T09:16:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:F5LM7OP2EO26MTKT34ITXHY35P","target":"record","payload":{"canonical_record":{"source":{"id":"2410.03907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T20:21:40Z","cross_cats_sorted":[],"title_canon_sha256":"94462c6328a1b78451e32a6ed8e09e99d2e2e0f0091c4b4d6867df568092a2b5","abstract_canon_sha256":"263ccda2da4aaf44d2156c6249561831c8a535ab4950f17237545ccf121dcbe8"},"schema_version":"1.0"},"canonical_sha256":"2f56cfb9fa23b5e64d53df113b9f1bebe21a49049a2debc1f304787a523a4852","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:26.272982Z","signature_b64":"oqXfYa7vjmix41USKULjjdyvnsD5bBLADItwsope9d9RP7gYwXYUtGoL5uR/jJxEi7bF420GgXD4i0rVp1JfDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f56cfb9fa23b5e64d53df113b9f1bebe21a49049a2debc1f304787a523a4852","last_reissued_at":"2026-07-05T09:16:26.272505Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:26.272505Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.03907","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":"2026-07-05T09:16:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OwbY3mvW6RXWxUvU+6jDJwd/hpHgFJV6Zcpd5d4aRlvrJGhT4pj4vMlAT3sB4MWYQLNRPYFEHakjvkHxf5CHAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:14:35.876059Z"},"content_sha256":"c7a8c0e4645eac51015e545ae08707feaa063ff39584c76020127d4411d2a514","schema_version":"1.0","event_id":"sha256:c7a8c0e4645eac51015e545ae08707feaa063ff39584c76020127d4411d2a514"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:F5LM7OP2EO26MTKT34ITXHY35P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ActPlan-1K: Benchmarking the Procedural Planning Ability of Visual Language Models in Household Activities","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haochen Shi, Jiayang Cheng, Yangqiu Song, Yauwai Yim, Ying Su, Zhan Ling","submitted_at":"2024-10-04T20:21:40Z","abstract_excerpt":"Large language models~(LLMs) have been adopted to process textual task description and accomplish procedural planning in embodied AI tasks because of their powerful reasoning ability. However, there is still lack of study on how vision language models~(VLMs) behave when multi-modal task inputs are considered. Counterfactual planning that evaluates the model's reasoning ability over alternative task situations are also under exploited. In order to evaluate the planning ability of both multi-modal and counterfactual aspects, we propose ActPlan-1K. ActPlan-1K is a multi-modal planning benchmark c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03907","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/2410.03907/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":"2026-07-05T09:16:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j1kJhBBsW+1tagfPq4f+0tPGotqMjnDfIw2gzvity/uUwOyUxtYFuHO6WXphvoTUYOiH/o6ErLkYiBo2+bxIAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:14:35.876692Z"},"content_sha256":"2a5647e3257550c50da209eeb7ca34338c2759292e4437b698ee454ab04dca3f","schema_version":"1.0","event_id":"sha256:2a5647e3257550c50da209eeb7ca34338c2759292e4437b698ee454ab04dca3f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/F5LM7OP2EO26MTKT34ITXHY35P/bundle.json","state_url":"https://pith.science/pith/F5LM7OP2EO26MTKT34ITXHY35P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/F5LM7OP2EO26MTKT34ITXHY35P/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":"2026-08-08T10:14:35Z","links":{"resolver":"https://pith.science/pith/F5LM7OP2EO26MTKT34ITXHY35P","bundle":"https://pith.science/pith/F5LM7OP2EO26MTKT34ITXHY35P/bundle.json","state":"https://pith.science/pith/F5LM7OP2EO26MTKT34ITXHY35P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/F5LM7OP2EO26MTKT34ITXHY35P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:F5LM7OP2EO26MTKT34ITXHY35P","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":"263ccda2da4aaf44d2156c6249561831c8a535ab4950f17237545ccf121dcbe8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T20:21:40Z","title_canon_sha256":"94462c6328a1b78451e32a6ed8e09e99d2e2e0f0091c4b4d6867df568092a2b5"},"schema_version":"1.0","source":{"id":"2410.03907","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03907","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03907v1","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03907","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_12","alias_value":"F5LM7OP2EO26","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_16","alias_value":"F5LM7OP2EO26MTKT","created_at":"2026-07-05T09:16:26Z"},{"alias_kind":"pith_short_8","alias_value":"F5LM7OP2","created_at":"2026-07-05T09:16:26Z"}],"graph_snapshots":[{"event_id":"sha256:2a5647e3257550c50da209eeb7ca34338c2759292e4437b698ee454ab04dca3f","target":"graph","created_at":"2026-07-05T09:16:26Z","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/2410.03907/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models~(LLMs) have been adopted to process textual task description and accomplish procedural planning in embodied AI tasks because of their powerful reasoning ability. However, there is still lack of study on how vision language models~(VLMs) behave when multi-modal task inputs are considered. Counterfactual planning that evaluates the model's reasoning ability over alternative task situations are also under exploited. In order to evaluate the planning ability of both multi-modal and counterfactual aspects, we propose ActPlan-1K. ActPlan-1K is a multi-modal planning benchmark c","authors_text":"Haochen Shi, Jiayang Cheng, Yangqiu Song, Yauwai Yim, Ying Su, Zhan Ling","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T20:21:40Z","title":"ActPlan-1K: Benchmarking the Procedural Planning Ability of Visual Language Models in Household Activities"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03907","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:c7a8c0e4645eac51015e545ae08707feaa063ff39584c76020127d4411d2a514","target":"record","created_at":"2026-07-05T09:16:26Z","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":"263ccda2da4aaf44d2156c6249561831c8a535ab4950f17237545ccf121dcbe8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T20:21:40Z","title_canon_sha256":"94462c6328a1b78451e32a6ed8e09e99d2e2e0f0091c4b4d6867df568092a2b5"},"schema_version":"1.0","source":{"id":"2410.03907","kind":"arxiv","version":1}},"canonical_sha256":"2f56cfb9fa23b5e64d53df113b9f1bebe21a49049a2debc1f304787a523a4852","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2f56cfb9fa23b5e64d53df113b9f1bebe21a49049a2debc1f304787a523a4852","first_computed_at":"2026-07-05T09:16:26.272505Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:16:26.272505Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oqXfYa7vjmix41USKULjjdyvnsD5bBLADItwsope9d9RP7gYwXYUtGoL5uR/jJxEi7bF420GgXD4i0rVp1JfDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:16:26.272982Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.03907","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c7a8c0e4645eac51015e545ae08707feaa063ff39584c76020127d4411d2a514","sha256:2a5647e3257550c50da209eeb7ca34338c2759292e4437b698ee454ab04dca3f"],"state_sha256":"117494bb1829ea4673d42649e749771b36f55d70bec86819fc21b9d0eff7df3a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RHnqM90sJ3bOVnw1jS14b4FVEPzK4TQO57RLQXszo/Q29pIKU/zZWdqlYS1YzhXJdmpVgiRNosD1WCvFWkWRDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T10:14:35.882975Z","bundle_sha256":"69ebab3ad67795e444a3ee6c57a7d3a138df2e043a32fb173d6e6c1b172e53b1"}}