{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:INAW6RHWQRW3PIL67MLK5VKRAP","short_pith_number":"pith:INAW6RHW","schema_version":"1.0","canonical_sha256":"43416f44f6846db7a17efb16aed55103cf0879e2aa707a1044ffb4521f759c8a","source":{"kind":"arxiv","id":"2606.08035","version":1},"attestation_state":"computed","paper":{"title":"DyCo-RL: Dynamic Cross-Modal Coordination for Visual Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chi Liu, Hangui Lin, Minghao Qin, Nicu Sebe, Teng Long, Xiangrui Liu, Yan Shu, Zheng Liu, Zhengyang Liang","submitted_at":"2026-06-06T07:56:25Z","abstract_excerpt":"Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a leading paradigm for enhancing visual reasoning in Multimodal Large Language Models (MLLMs). However, existing RLVR methods optimize primarily for the reasoning outcome, fundamentally overlooking the fine-grained cross-modal coordination required during the generation process. Through token-level analyses and controlled interventions, we reveal that during Chain-of-Thought (CoT) reasoning, MLLMs frequently fail to dynamically alternate between extracting visual evidence and synthesizing textual context-a coordination breakd"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.08035","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-06T07:56:25Z","cross_cats_sorted":[],"title_canon_sha256":"eb20e9bab27d5485735de9481643bec9684ab8445121c9bd5225a1ad81496614","abstract_canon_sha256":"8ec255e3ed0df4581187d928a79c1469ae46c2e02c0091475d5e8c7f956486a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:23.813455Z","signature_b64":"Jq0oGB1ABIqdgCv3v1Y63ERviKqj0m9lStQba5zR4Kfwc4DDbqRcv59ea8X1VYgyNy9bDT9HD4o3Z5suxvzrAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43416f44f6846db7a17efb16aed55103cf0879e2aa707a1044ffb4521f759c8a","last_reissued_at":"2026-06-09T01:05:23.812760Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:23.812760Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DyCo-RL: Dynamic Cross-Modal Coordination for Visual Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chi Liu, Hangui Lin, Minghao Qin, Nicu Sebe, Teng Long, Xiangrui Liu, Yan Shu, Zheng Liu, Zhengyang Liang","submitted_at":"2026-06-06T07:56:25Z","abstract_excerpt":"Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a leading paradigm for enhancing visual reasoning in Multimodal Large Language Models (MLLMs). However, existing RLVR methods optimize primarily for the reasoning outcome, fundamentally overlooking the fine-grained cross-modal coordination required during the generation process. Through token-level analyses and controlled interventions, we reveal that during Chain-of-Thought (CoT) reasoning, MLLMs frequently fail to dynamically alternate between extracting visual evidence and synthesizing textual context-a coordination breakd"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08035","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/2606.08035/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.08035","created_at":"2026-06-09T01:05:23.812866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08035v1","created_at":"2026-06-09T01:05:23.812866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08035","created_at":"2026-06-09T01:05:23.812866+00:00"},{"alias_kind":"pith_short_12","alias_value":"INAW6RHWQRW3","created_at":"2026-06-09T01:05:23.812866+00:00"},{"alias_kind":"pith_short_16","alias_value":"INAW6RHWQRW3PIL6","created_at":"2026-06-09T01:05:23.812866+00:00"},{"alias_kind":"pith_short_8","alias_value":"INAW6RHW","created_at":"2026-06-09T01:05:23.812866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP","json":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP.json","graph_json":"https://pith.science/api/pith-number/INAW6RHWQRW3PIL67MLK5VKRAP/graph.json","events_json":"https://pith.science/api/pith-number/INAW6RHWQRW3PIL67MLK5VKRAP/events.json","paper":"https://pith.science/paper/INAW6RHW"},"agent_actions":{"view_html":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP","download_json":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP.json","view_paper":"https://pith.science/paper/INAW6RHW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08035&json=true","fetch_graph":"https://pith.science/api/pith-number/INAW6RHWQRW3PIL67MLK5VKRAP/graph.json","fetch_events":"https://pith.science/api/pith-number/INAW6RHWQRW3PIL67MLK5VKRAP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP/action/storage_attestation","attest_author":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP/action/author_attestation","sign_citation":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP/action/citation_signature","submit_replication":"https://pith.science/pith/INAW6RHWQRW3PIL67MLK5VKRAP/action/replication_record"}},"created_at":"2026-06-09T01:05:23.812866+00:00","updated_at":"2026-06-09T01:05:23.812866+00:00"}