{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:EFP2DXNGKEE7KAXDF2P2GW3VID","short_pith_number":"pith:EFP2DXNG","canonical_record":{"source":{"id":"2503.23064","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-29T12:50:38Z","cross_cats_sorted":[],"title_canon_sha256":"209330c47888af97b07edc4094805c91bf8059f19c76dd698efb5d945055ecee","abstract_canon_sha256":"f6b50b012468001ef21fcf28c0b5f4d3bb6faeee3212f735ae54d843a09e0e1a"},"schema_version":"1.0"},"canonical_sha256":"215fa1dda65109f502e32e9fa35b7540d4ea2b42b8e66638c14695bbdcb02364","source":{"kind":"arxiv","id":"2503.23064","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.23064","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"arxiv_version","alias_value":"2503.23064v2","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.23064","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_12","alias_value":"EFP2DXNGKEE7","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_16","alias_value":"EFP2DXNGKEE7KAXD","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_8","alias_value":"EFP2DXNG","created_at":"2026-07-05T10:43:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:EFP2DXNGKEE7KAXDF2P2GW3VID","target":"record","payload":{"canonical_record":{"source":{"id":"2503.23064","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-29T12:50:38Z","cross_cats_sorted":[],"title_canon_sha256":"209330c47888af97b07edc4094805c91bf8059f19c76dd698efb5d945055ecee","abstract_canon_sha256":"f6b50b012468001ef21fcf28c0b5f4d3bb6faeee3212f735ae54d843a09e0e1a"},"schema_version":"1.0"},"canonical_sha256":"215fa1dda65109f502e32e9fa35b7540d4ea2b42b8e66638c14695bbdcb02364","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:07.082949Z","signature_b64":"yWCLnFdQdpAKi5C5H7gvayzAYT7+neOw9bTUZgS4Zi0XPhM8a8OLIeYh2wD21Z4f6frrJ/LMnGiHHLgFSNq8CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"215fa1dda65109f502e32e9fa35b7540d4ea2b42b8e66638c14695bbdcb02364","last_reissued_at":"2026-07-05T10:43:07.082447Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:07.082447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.23064","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:43:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DwmscpQN299IsSzIti+UKV2af37d84oCpTD0fkTqvp9HUZ5/W2HS8BV5T6XJdi5VLG443ry7R3TwjrQm82aBCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T23:27:21.738500Z"},"content_sha256":"be6209cd2996eef3582b2cbe3e89be4eeb901b63f391962a7f2b858366b2ca28","schema_version":"1.0","event_id":"sha256:be6209cd2996eef3582b2cbe3e89be4eeb901b63f391962a7f2b858366b2ca28"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:EFP2DXNGKEE7KAXDF2P2GW3VID","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"VGRP-Bench: Visual Grid Reasoning Puzzle Benchmark for Large Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Filippos Kokkinos, Junlin Han, Konstantinos Tertikas, Sabine S\\\"usstrunk, Shalini Maiti, Tong Zhang, Yufan Ren","submitted_at":"2025-03-29T12:50:38Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) struggle with puzzles, which require precise perception, rule comprehension, and logical reasoning. Assessing and enhancing their performance in this domain is crucial, as it reflects their ability to engage in structured reasoning - an essential skill for real-world problem-solving. However, existing benchmarks primarily evaluate pre-trained models without additional training or fine-tuning, often lack a dedicated focus on reasoning, and fail to establish a systematic evaluation framework. To address these limitations, we introduce VGRP-Bench, a Visual Gri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.23064","kind":"arxiv","version":2},"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/2503.23064/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":"2026-07-05T10:43:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vGLqIFfNJ04cJIQ+k70r8lehiDBnkAc22mx4/52Z8NJDQYWH9/Yy4ehY0baMu+j9hp3XdQBo5KyQQm2PBU66Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T23:27:21.739000Z"},"content_sha256":"5f16dfb40a060ac6c98682ddff7cd284b81e71a8b6ed2d1ac930b068e4244687","schema_version":"1.0","event_id":"sha256:5f16dfb40a060ac6c98682ddff7cd284b81e71a8b6ed2d1ac930b068e4244687"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/bundle.json","state_url":"https://pith.science/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/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":"2026-08-05T23:27:21Z","links":{"resolver":"https://pith.science/pith/EFP2DXNGKEE7KAXDF2P2GW3VID","bundle":"https://pith.science/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/bundle.json","state":"https://pith.science/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EFP2DXNGKEE7KAXDF2P2GW3VID/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EFP2DXNGKEE7KAXDF2P2GW3VID","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":"f6b50b012468001ef21fcf28c0b5f4d3bb6faeee3212f735ae54d843a09e0e1a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-29T12:50:38Z","title_canon_sha256":"209330c47888af97b07edc4094805c91bf8059f19c76dd698efb5d945055ecee"},"schema_version":"1.0","source":{"id":"2503.23064","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.23064","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"arxiv_version","alias_value":"2503.23064v2","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.23064","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_12","alias_value":"EFP2DXNGKEE7","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_16","alias_value":"EFP2DXNGKEE7KAXD","created_at":"2026-07-05T10:43:07Z"},{"alias_kind":"pith_short_8","alias_value":"EFP2DXNG","created_at":"2026-07-05T10:43:07Z"}],"graph_snapshots":[{"event_id":"sha256:5f16dfb40a060ac6c98682ddff7cd284b81e71a8b6ed2d1ac930b068e4244687","target":"graph","created_at":"2026-07-05T10:43:07Z","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/2503.23064/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Vision-Language Models (LVLMs) struggle with puzzles, which require precise perception, rule comprehension, and logical reasoning. Assessing and enhancing their performance in this domain is crucial, as it reflects their ability to engage in structured reasoning - an essential skill for real-world problem-solving. However, existing benchmarks primarily evaluate pre-trained models without additional training or fine-tuning, often lack a dedicated focus on reasoning, and fail to establish a systematic evaluation framework. To address these limitations, we introduce VGRP-Bench, a Visual Gri","authors_text":"Filippos Kokkinos, Junlin Han, Konstantinos Tertikas, Sabine S\\\"usstrunk, Shalini Maiti, Tong Zhang, Yufan Ren","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-29T12:50:38Z","title":"VGRP-Bench: Visual Grid Reasoning Puzzle Benchmark for Large Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.23064","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:be6209cd2996eef3582b2cbe3e89be4eeb901b63f391962a7f2b858366b2ca28","target":"record","created_at":"2026-07-05T10:43:07Z","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":"f6b50b012468001ef21fcf28c0b5f4d3bb6faeee3212f735ae54d843a09e0e1a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-29T12:50:38Z","title_canon_sha256":"209330c47888af97b07edc4094805c91bf8059f19c76dd698efb5d945055ecee"},"schema_version":"1.0","source":{"id":"2503.23064","kind":"arxiv","version":2}},"canonical_sha256":"215fa1dda65109f502e32e9fa35b7540d4ea2b42b8e66638c14695bbdcb02364","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"215fa1dda65109f502e32e9fa35b7540d4ea2b42b8e66638c14695bbdcb02364","first_computed_at":"2026-07-05T10:43:07.082447Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:43:07.082447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yWCLnFdQdpAKi5C5H7gvayzAYT7+neOw9bTUZgS4Zi0XPhM8a8OLIeYh2wD21Z4f6frrJ/LMnGiHHLgFSNq8CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:43:07.082949Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.23064","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be6209cd2996eef3582b2cbe3e89be4eeb901b63f391962a7f2b858366b2ca28","sha256:5f16dfb40a060ac6c98682ddff7cd284b81e71a8b6ed2d1ac930b068e4244687"],"state_sha256":"d7e7cd6e254d1ddfca2485fd79d04efd5cc1c4b01cf4887da1558bd79acafd52"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2KxroCBdiKu/l4mySQ22/N8Qz0awZRtuVvqrRr194/m+4qx95vlNZsi+oE6oeFGkG2IWsXP/Tr/LXjztX5mIAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T23:27:21.742850Z","bundle_sha256":"4092749f1f4a124865a932aa2f1d4f2bcfb76f1f511c179fd6197ab5121e6511"}}