{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:L4NCZO3QI4BLHFKQ7UB33NSFG4","short_pith_number":"pith:L4NCZO3Q","canonical_record":{"source":{"id":"2608.13267","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-13T14:06:35Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"2259f25f54db20a4d706a3466c0f4f9a026fca8fab83cbd3bd3a287391e7246f","abstract_canon_sha256":"7d363ed63d31f2229f815b49beb50166dcd7f578eba9c230ec3ad86b11189193"},"schema_version":"1.0"},"canonical_sha256":"5f1a2cbb704702b39550fd03bdb64537122d5bbbd75ca6d988294992c55085bd","source":{"kind":"arxiv","id":"2608.13267","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13267","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13267v1","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13267","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_12","alias_value":"L4NCZO3QI4BL","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_16","alias_value":"L4NCZO3QI4BLHFKQ","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_8","alias_value":"L4NCZO3Q","created_at":"2026-08-14T01:00:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:L4NCZO3QI4BLHFKQ7UB33NSFG4","target":"record","payload":{"canonical_record":{"source":{"id":"2608.13267","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-13T14:06:35Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"2259f25f54db20a4d706a3466c0f4f9a026fca8fab83cbd3bd3a287391e7246f","abstract_canon_sha256":"7d363ed63d31f2229f815b49beb50166dcd7f578eba9c230ec3ad86b11189193"},"schema_version":"1.0"},"canonical_sha256":"5f1a2cbb704702b39550fd03bdb64537122d5bbbd75ca6d988294992c55085bd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-14T01:00:49.074121Z","signature_b64":"0m+rGdYnBchWuAA8H+9UQcem160upAqLusrMCaA4KMG6c5JwO5piDQ9Twki5cLFpRODSI+G7x3ILy1iCr5AXDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f1a2cbb704702b39550fd03bdb64537122d5bbbd75ca6d988294992c55085bd","last_reissued_at":"2026-08-14T01:00:49.072027Z","signature_status":"signed_v1","first_computed_at":"2026-08-14T01:00:49.072027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.13267","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":"2026-08-14T01:00:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FlGE0Xi4ovDxQbVmAhtwq5OD77R88IDqp2T1W49YwiXyHrX2/EgSoqZywHO8e4FAiFxRygFJriLoo4cNBpLwCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T11:33:57.812938Z"},"content_sha256":"20f9a8b86058d37795ae2b5eb2f104172415c78a92a5463b089e17a9bf0bb968","schema_version":"1.0","event_id":"sha256:20f9a8b86058d37795ae2b5eb2f104172415c78a92a5463b089e17a9bf0bb968"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:L4NCZO3QI4BLHFKQ7UB33NSFG4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Do VLMs Behave When Blind or Misled? Behavioral Evaluation of VLMs on Scientific Figures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ananya Mukherjee, Christian Greisinger, Owusu-Banahene Osei, Paul Osemudiame Oamen, Pius Onobhayedo, Steffen Eger, Wei Zhao","submitted_at":"2026-08-13T14:06:35Z","abstract_excerpt":"Existing vision-language model (VLM) benchmarks emphasize perception and reasoning accuracy (how well VLMs describe and reason about what they see in an image), with limited attention to behavioral reliability under uncertainty (how they behave when visual evidence is missing or misleading). We introduce SciFigBench, a diagnostic VLM benchmark for scientific figure understanding that jointly evaluates perception, reasoning, and behavioral reliability under uncertainty. It contains 250 figures with high-quality human annotations across three evaluation aspects, totaling 600+ hours of annotation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13267","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/2608.13267/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":"2026-08-14T01:00:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uUrNE4YTz9Szn0XZ5WNKdYrnYYHormcGGEOZgJnp9bhEAQg1XipwQmf+IiIlRLP3k4r3SLD8qlZReSvNJaWHCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T11:33:57.813883Z"},"content_sha256":"2dd0f7e580e1a9d29c2063fadba0234cb95609c9dd0ef943202a533e93e290f0","schema_version":"1.0","event_id":"sha256:2dd0f7e580e1a9d29c2063fadba0234cb95609c9dd0ef943202a533e93e290f0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/bundle.json","state_url":"https://pith.science/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/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":"2026-08-16T11:33:57Z","links":{"resolver":"https://pith.science/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4","bundle":"https://pith.science/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/bundle.json","state":"https://pith.science/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L4NCZO3QI4BLHFKQ7UB33NSFG4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:L4NCZO3QI4BLHFKQ7UB33NSFG4","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":"7d363ed63d31f2229f815b49beb50166dcd7f578eba9c230ec3ad86b11189193","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-13T14:06:35Z","title_canon_sha256":"2259f25f54db20a4d706a3466c0f4f9a026fca8fab83cbd3bd3a287391e7246f"},"schema_version":"1.0","source":{"id":"2608.13267","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13267","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13267v1","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13267","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_12","alias_value":"L4NCZO3QI4BL","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_16","alias_value":"L4NCZO3QI4BLHFKQ","created_at":"2026-08-14T01:00:49Z"},{"alias_kind":"pith_short_8","alias_value":"L4NCZO3Q","created_at":"2026-08-14T01:00:49Z"}],"graph_snapshots":[{"event_id":"sha256:2dd0f7e580e1a9d29c2063fadba0234cb95609c9dd0ef943202a533e93e290f0","target":"graph","created_at":"2026-08-14T01:00:49Z","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/2608.13267/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing vision-language model (VLM) benchmarks emphasize perception and reasoning accuracy (how well VLMs describe and reason about what they see in an image), with limited attention to behavioral reliability under uncertainty (how they behave when visual evidence is missing or misleading). We introduce SciFigBench, a diagnostic VLM benchmark for scientific figure understanding that jointly evaluates perception, reasoning, and behavioral reliability under uncertainty. It contains 250 figures with high-quality human annotations across three evaluation aspects, totaling 600+ hours of annotation","authors_text":"Ananya Mukherjee, Christian Greisinger, Owusu-Banahene Osei, Paul Osemudiame Oamen, Pius Onobhayedo, Steffen Eger, Wei Zhao","cross_cats":["cs.AI","cs.CV","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-13T14:06:35Z","title":"How Do VLMs Behave When Blind or Misled? Behavioral Evaluation of VLMs on Scientific Figures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13267","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":"sha256:20f9a8b86058d37795ae2b5eb2f104172415c78a92a5463b089e17a9bf0bb968","target":"record","created_at":"2026-08-14T01:00:49Z","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":"7d363ed63d31f2229f815b49beb50166dcd7f578eba9c230ec3ad86b11189193","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-13T14:06:35Z","title_canon_sha256":"2259f25f54db20a4d706a3466c0f4f9a026fca8fab83cbd3bd3a287391e7246f"},"schema_version":"1.0","source":{"id":"2608.13267","kind":"arxiv","version":1}},"canonical_sha256":"5f1a2cbb704702b39550fd03bdb64537122d5bbbd75ca6d988294992c55085bd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5f1a2cbb704702b39550fd03bdb64537122d5bbbd75ca6d988294992c55085bd","first_computed_at":"2026-08-14T01:00:49.072027Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-14T01:00:49.072027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0m+rGdYnBchWuAA8H+9UQcem160upAqLusrMCaA4KMG6c5JwO5piDQ9Twki5cLFpRODSI+G7x3ILy1iCr5AXDw==","signature_status":"signed_v1","signed_at":"2026-08-14T01:00:49.074121Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.13267","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:20f9a8b86058d37795ae2b5eb2f104172415c78a92a5463b089e17a9bf0bb968","sha256:2dd0f7e580e1a9d29c2063fadba0234cb95609c9dd0ef943202a533e93e290f0"],"state_sha256":"307488f19f9113b5bbe1fa4ee18d3387d0db9f074c8de1b950845aab62ca12b7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GQ8WYEDbG7vdzI7Q/FHICqQvKVDjmIW6yScpVv1yGCRjThsPucfqbDo0UPvdhH96jxjKv0ywTnyCBvik1R0fDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T11:33:57.820102Z","bundle_sha256":"a45f89e269cd71846a40a63bd318d8a1f5f0d0730877570f5a2973fd56c1f1fc"}}