{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:36COOF52A5Y3C4IZPN2POEOZLW","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":"91a234ba7ad34b1f3d737158b9307cce512bf67e8db8669b7b7f34e2d36fc52b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T02:02:23Z","title_canon_sha256":"9fa27e6474ddfaf381990dd96e46a5763c13e57b11c03f15485aec287c24941e"},"schema_version":"1.0","source":{"id":"2606.01558","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.01558","created_at":"2026-06-02T02:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2606.01558v1","created_at":"2026-06-02T02:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01558","created_at":"2026-06-02T02:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"36COOF52A5Y3","created_at":"2026-06-02T02:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"36COOF52A5Y3C4IZ","created_at":"2026-06-02T02:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"36COOF52","created_at":"2026-06-02T02:04:36Z"}],"graph_snapshots":[{"event_id":"sha256:7a5898df1ba97bb883e9cc9ce45201252d9808f0817014e785808853a8d88ae5","target":"graph","created_at":"2026-06-02T02:04:36Z","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/2606.01558/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The effectiveness of Chain-of-Thought (CoT) prompting in Multimodal Large Language Models (MLLMs) remains uncertain: across several visual reasoning benchmarks, CoT prompting often degrades performance compared to direct prompting. In this paper, we provide a systematic analysis of CoT behavior in three modern MLLM families across model scales on datasets requiring step-wise visual evidence. Our analysis identifies two recurring failure modes: premature answer commitment and limited direct visual-token access during rationale generation. We further find that standard CoT-style Supervised Fine-","authors_text":"Aidong Zhang, Bohan Liu, Guangzhi Xiong, Sanchit Sinha, Zhenghao He","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T02:02:23Z","title":"Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01558","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":"sha256:4fcfb35a8c34133175434b0f2fbbaf088120bd80e0a1d796887bd9f0a8b8a2b0","target":"record","created_at":"2026-06-02T02:04:36Z","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":"91a234ba7ad34b1f3d737158b9307cce512bf67e8db8669b7b7f34e2d36fc52b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T02:02:23Z","title_canon_sha256":"9fa27e6474ddfaf381990dd96e46a5763c13e57b11c03f15485aec287c24941e"},"schema_version":"1.0","source":{"id":"2606.01558","kind":"arxiv","version":1}},"canonical_sha256":"df84e717ba0771b171197b74f711d95d870160024c021ec82b34294b515aac60","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"df84e717ba0771b171197b74f711d95d870160024c021ec82b34294b515aac60","first_computed_at":"2026-06-02T02:04:36.458451Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:36.458451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qoekP7EcKJHh/yBpUSFl0IBUFE/Ht4OTCjbHA5KjLznjQumXXqo/cvpK/58VMsKrret2vPA+RQ9Sj1yJnoI5DQ==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:36.458866Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.01558","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4fcfb35a8c34133175434b0f2fbbaf088120bd80e0a1d796887bd9f0a8b8a2b0","sha256:7a5898df1ba97bb883e9cc9ce45201252d9808f0817014e785808853a8d88ae5"],"state_sha256":"f27bb2b69081f89cd0930eeac9f43047d384f8a7b5745129788d065b6a01ca76"}