{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Q3RSAZ6KLGETVXDJSMGTWGJYPJ","short_pith_number":"pith:Q3RSAZ6K","canonical_record":{"source":{"id":"2606.00148","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T03:20:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"be9c9d8216998d2c916e1c78acf1464908f234f56549d500e660116643616b32","abstract_canon_sha256":"aa4c3ee15b1dd83c85b4ea4a28ae0dda92650a85cb94686ec50a663744a5048c"},"schema_version":"1.0"},"canonical_sha256":"86e32067ca59893adc69930d3b19387a516661ba0dad9816325c7114a496b551","source":{"kind":"arxiv","id":"2606.00148","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00148","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00148v1","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00148","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_12","alias_value":"Q3RSAZ6KLGET","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_16","alias_value":"Q3RSAZ6KLGETVXDJ","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_8","alias_value":"Q3RSAZ6K","created_at":"2026-06-02T01:03:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Q3RSAZ6KLGETVXDJSMGTWGJYPJ","target":"record","payload":{"canonical_record":{"source":{"id":"2606.00148","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T03:20:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"be9c9d8216998d2c916e1c78acf1464908f234f56549d500e660116643616b32","abstract_canon_sha256":"aa4c3ee15b1dd83c85b4ea4a28ae0dda92650a85cb94686ec50a663744a5048c"},"schema_version":"1.0"},"canonical_sha256":"86e32067ca59893adc69930d3b19387a516661ba0dad9816325c7114a496b551","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:19.819278Z","signature_b64":"yUa4r6KhiFsD8O8ZYGdUjHj7bxg3kultPf9hWlKqkf4xekiPutoWG176JwpzceicYEDvjnyGW4UUN9Y9yfNaAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86e32067ca59893adc69930d3b19387a516661ba0dad9816325c7114a496b551","last_reissued_at":"2026-06-02T01:03:19.818871Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:19.818871Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.00148","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":"2026-06-02T01:03:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tht6MdfcqCniic8mJ3GbODsIauu5e5DoufrrgVeBl6X7ajHoGnnVZCpjmcw1YACGAyZzZZ6Pd4XSWFuPgijODg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T00:43:10.612676Z"},"content_sha256":"e8947ba72b12976b23eda0b85e74153503bbf57b244c312ae1c36ab60bcafbe5","schema_version":"1.0","event_id":"sha256:e8947ba72b12976b23eda0b85e74153503bbf57b244c312ae1c36ab60bcafbe5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Q3RSAZ6KLGETVXDJSMGTWGJYPJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ao Cheng, Baiqi Wu, Qingyong Hu, Qiyao Sun, Xingming Li, Xixiang He, Xuanyu Ji","submitted_at":"2026-05-29T03:20:05Z","abstract_excerpt":"Multimodal large language models (MLLMs) often know the rule but pick the wrong answer: on abstract visual reasoning (AVR) tasks, a model can describe what it sees and name the underlying pattern, yet still fail to choose the matching candidate. Existing AVR benchmarks cannot detect this because they collapse perception, rule induction, and answer selection into a single right-or-wrong signal. We introduce StemBind, a shared-stem diagnostic benchmark that probes the same visual stem with three aligned questions: Perception (what is in the image), Rule (what pattern governs it), and Full (which"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00148","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/2606.00148/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":"2026-06-02T01:03:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pEPxt2wR7Iyzvma8fgnIB7GQNjYmfNrIP6N4SnZ4M9XKLWLI6s480Vs0aOFs5F7qdkgugmiKBwLFRyaPsPdMDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T00:43:10.613205Z"},"content_sha256":"325a0e3884724f2383dd63c48f661b6cbe1990e215a0140cfa72ef14f992084b","schema_version":"1.0","event_id":"sha256:325a0e3884724f2383dd63c48f661b6cbe1990e215a0140cfa72ef14f992084b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/bundle.json","state_url":"https://pith.science/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/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":"2026-06-11T00:43:10Z","links":{"resolver":"https://pith.science/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ","bundle":"https://pith.science/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/bundle.json","state":"https://pith.science/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q3RSAZ6KLGETVXDJSMGTWGJYPJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Q3RSAZ6KLGETVXDJSMGTWGJYPJ","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":"aa4c3ee15b1dd83c85b4ea4a28ae0dda92650a85cb94686ec50a663744a5048c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T03:20:05Z","title_canon_sha256":"be9c9d8216998d2c916e1c78acf1464908f234f56549d500e660116643616b32"},"schema_version":"1.0","source":{"id":"2606.00148","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00148","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00148v1","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00148","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_12","alias_value":"Q3RSAZ6KLGET","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_16","alias_value":"Q3RSAZ6KLGETVXDJ","created_at":"2026-06-02T01:03:19Z"},{"alias_kind":"pith_short_8","alias_value":"Q3RSAZ6K","created_at":"2026-06-02T01:03:19Z"}],"graph_snapshots":[{"event_id":"sha256:325a0e3884724f2383dd63c48f661b6cbe1990e215a0140cfa72ef14f992084b","target":"graph","created_at":"2026-06-02T01:03:19Z","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/2606.00148/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal large language models (MLLMs) often know the rule but pick the wrong answer: on abstract visual reasoning (AVR) tasks, a model can describe what it sees and name the underlying pattern, yet still fail to choose the matching candidate. Existing AVR benchmarks cannot detect this because they collapse perception, rule induction, and answer selection into a single right-or-wrong signal. We introduce StemBind, a shared-stem diagnostic benchmark that probes the same visual stem with three aligned questions: Perception (what is in the image), Rule (what pattern governs it), and Full (which","authors_text":"Ao Cheng, Baiqi Wu, Qingyong Hu, Qiyao Sun, Xingming Li, Xixiang He, Xuanyu Ji","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T03:20:05Z","title":"StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00148","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:e8947ba72b12976b23eda0b85e74153503bbf57b244c312ae1c36ab60bcafbe5","target":"record","created_at":"2026-06-02T01:03:19Z","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":"aa4c3ee15b1dd83c85b4ea4a28ae0dda92650a85cb94686ec50a663744a5048c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T03:20:05Z","title_canon_sha256":"be9c9d8216998d2c916e1c78acf1464908f234f56549d500e660116643616b32"},"schema_version":"1.0","source":{"id":"2606.00148","kind":"arxiv","version":1}},"canonical_sha256":"86e32067ca59893adc69930d3b19387a516661ba0dad9816325c7114a496b551","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"86e32067ca59893adc69930d3b19387a516661ba0dad9816325c7114a496b551","first_computed_at":"2026-06-02T01:03:19.818871Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:19.818871Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yUa4r6KhiFsD8O8ZYGdUjHj7bxg3kultPf9hWlKqkf4xekiPutoWG176JwpzceicYEDvjnyGW4UUN9Y9yfNaAg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:19.819278Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00148","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e8947ba72b12976b23eda0b85e74153503bbf57b244c312ae1c36ab60bcafbe5","sha256:325a0e3884724f2383dd63c48f661b6cbe1990e215a0140cfa72ef14f992084b"],"state_sha256":"9e0295460eb017d4bc2520fa0521e1527022897932fed82ec7718610de72d401"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mLFaJgsUhy+pQMWGjhc2LCCiAvn2fMj3x1rchlp908/HMdONVbt02ukA4Y57tv/601JJ1k7VvqaoEToxf88PBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T00:43:10.617784Z","bundle_sha256":"428bf1e68d5a61cb9715d94e75305ff7198d7aca122e95ec507886e4a51f1935"}}