{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:MAE7GM5SCWK4SPNFZWEBW3XIOR","short_pith_number":"pith:MAE7GM5S","canonical_record":{"source":{"id":"2410.06555","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T05:17:38Z","cross_cats_sorted":[],"title_canon_sha256":"bfecb1d8b2dfa2c7c385073ba4767ede419426c2e89669be9431428e1d87214d","abstract_canon_sha256":"580e1a1a32c6d49fe9ec4a1029f0e4c493db3c583b9f25df91226814fc502a29"},"schema_version":"1.0"},"canonical_sha256":"6009f333b21595c93da5cd881b6ee874512af6404779bb2e79d1de8670f24af3","source":{"kind":"arxiv","id":"2410.06555","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.06555","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"arxiv_version","alias_value":"2410.06555v1","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.06555","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_12","alias_value":"MAE7GM5SCWK4","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_16","alias_value":"MAE7GM5SCWK4SPNF","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_8","alias_value":"MAE7GM5S","created_at":"2026-07-05T09:18:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:MAE7GM5SCWK4SPNFZWEBW3XIOR","target":"record","payload":{"canonical_record":{"source":{"id":"2410.06555","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T05:17:38Z","cross_cats_sorted":[],"title_canon_sha256":"bfecb1d8b2dfa2c7c385073ba4767ede419426c2e89669be9431428e1d87214d","abstract_canon_sha256":"580e1a1a32c6d49fe9ec4a1029f0e4c493db3c583b9f25df91226814fc502a29"},"schema_version":"1.0"},"canonical_sha256":"6009f333b21595c93da5cd881b6ee874512af6404779bb2e79d1de8670f24af3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:04.003707Z","signature_b64":"CxyFY+f2w5riUhS7ehzd5WK0A+xkXf3Zktaswh6Y58P8RFnoWT+QHhVJ3d4Anb2tGQQuikGTTQtw6j4T0z/DBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6009f333b21595c93da5cd881b6ee874512af6404779bb2e79d1de8670f24af3","last_reissued_at":"2026-07-05T09:18:04.003294Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:04.003294Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.06555","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":"2026-07-05T09:18:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bMD8n50/f9cH0ugq+SdJ971PUYrdVjZaNgHleJiOIdrdrJPkqtIwHnf0AtT1Vz/8ExKXeq2RC9Rn9874qF4xAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:20:01.346391Z"},"content_sha256":"25d5efd16307cb7fddbd01e1486012a01381a85e808ed3331ac23aa4f1fb50ed","schema_version":"1.0","event_id":"sha256:25d5efd16307cb7fddbd01e1486012a01381a85e808ed3331ac23aa4f1fb50ed"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:MAE7GM5SCWK4SPNFZWEBW3XIOR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ING-VP: MLLMs cannot Play Easy Vision-based Games Yet","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ge Zhang, Hangyu Guo, Haoran Zhang, Jiaheng Liu, Meng Cao, Shuyue Guo, Wenhao Huang","submitted_at":"2024-10-09T05:17:38Z","abstract_excerpt":"As multimodal large language models (MLLMs) continue to demonstrate increasingly competitive performance across a broad spectrum of tasks, more intricate and comprehensive benchmarks have been developed to assess these cutting-edge models. These benchmarks introduce new challenges to core capabilities such as perception, reasoning, and planning. However, existing multimodal benchmarks fall short in providing a focused evaluation of multi-step planning based on spatial relationships in images. To bridge this gap, we present ING-VP, the first INteractive Game-based Vision Planning benchmark, spe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.06555","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/2410.06555/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":"2026-07-05T09:18:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7S0lWqagvWktejnJKWU/oytpAJeAd53pyeR+MjizotedDwvgczpEj012ipYbvS6hRaR8S1gdbzJdgE3sjkjCAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:20:01.346918Z"},"content_sha256":"7d7fceb082bcaee4b9c95205eea4496627672374cac956efcc45ae5b4450bbad","schema_version":"1.0","event_id":"sha256:7d7fceb082bcaee4b9c95205eea4496627672374cac956efcc45ae5b4450bbad"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/bundle.json","state_url":"https://pith.science/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/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":"2026-08-08T18:20:01Z","links":{"resolver":"https://pith.science/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR","bundle":"https://pith.science/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/bundle.json","state":"https://pith.science/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MAE7GM5SCWK4SPNFZWEBW3XIOR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:MAE7GM5SCWK4SPNFZWEBW3XIOR","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":"580e1a1a32c6d49fe9ec4a1029f0e4c493db3c583b9f25df91226814fc502a29","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T05:17:38Z","title_canon_sha256":"bfecb1d8b2dfa2c7c385073ba4767ede419426c2e89669be9431428e1d87214d"},"schema_version":"1.0","source":{"id":"2410.06555","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.06555","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"arxiv_version","alias_value":"2410.06555v1","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.06555","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_12","alias_value":"MAE7GM5SCWK4","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_16","alias_value":"MAE7GM5SCWK4SPNF","created_at":"2026-07-05T09:18:04Z"},{"alias_kind":"pith_short_8","alias_value":"MAE7GM5S","created_at":"2026-07-05T09:18:04Z"}],"graph_snapshots":[{"event_id":"sha256:7d7fceb082bcaee4b9c95205eea4496627672374cac956efcc45ae5b4450bbad","target":"graph","created_at":"2026-07-05T09:18:04Z","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/2410.06555/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As multimodal large language models (MLLMs) continue to demonstrate increasingly competitive performance across a broad spectrum of tasks, more intricate and comprehensive benchmarks have been developed to assess these cutting-edge models. These benchmarks introduce new challenges to core capabilities such as perception, reasoning, and planning. However, existing multimodal benchmarks fall short in providing a focused evaluation of multi-step planning based on spatial relationships in images. To bridge this gap, we present ING-VP, the first INteractive Game-based Vision Planning benchmark, spe","authors_text":"Ge Zhang, Hangyu Guo, Haoran Zhang, Jiaheng Liu, Meng Cao, Shuyue Guo, Wenhao Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T05:17:38Z","title":"ING-VP: MLLMs cannot Play Easy Vision-based Games Yet"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.06555","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":"sha256:25d5efd16307cb7fddbd01e1486012a01381a85e808ed3331ac23aa4f1fb50ed","target":"record","created_at":"2026-07-05T09:18:04Z","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":"580e1a1a32c6d49fe9ec4a1029f0e4c493db3c583b9f25df91226814fc502a29","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T05:17:38Z","title_canon_sha256":"bfecb1d8b2dfa2c7c385073ba4767ede419426c2e89669be9431428e1d87214d"},"schema_version":"1.0","source":{"id":"2410.06555","kind":"arxiv","version":1}},"canonical_sha256":"6009f333b21595c93da5cd881b6ee874512af6404779bb2e79d1de8670f24af3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6009f333b21595c93da5cd881b6ee874512af6404779bb2e79d1de8670f24af3","first_computed_at":"2026-07-05T09:18:04.003294Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:18:04.003294Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CxyFY+f2w5riUhS7ehzd5WK0A+xkXf3Zktaswh6Y58P8RFnoWT+QHhVJ3d4Anb2tGQQuikGTTQtw6j4T0z/DBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:18:04.003707Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.06555","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:25d5efd16307cb7fddbd01e1486012a01381a85e808ed3331ac23aa4f1fb50ed","sha256:7d7fceb082bcaee4b9c95205eea4496627672374cac956efcc45ae5b4450bbad"],"state_sha256":"e99bb6d4d4e26f60fd74a2d61db428b769e66fb4daa76ea2761318eacf9c504c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FxO6PNudeIWcOrxw9KyiBMhlPCE+syTLUnxLCZNYvbNwQahZqXfD7TwiGV1MLPFDcj07yvDJYxJXMEgo246VDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T18:20:01.353192Z","bundle_sha256":"3dd5a9cfa993ff26a2c89912f31c820d809c29bbebb5f125f95959b39f83a1ae"}}