{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HI5DXARVG3UZXAXXSNNJVFBLF6","short_pith_number":"pith:HI5DXARV","schema_version":"1.0","canonical_sha256":"3a3a3b823536e99b82f7935a9a942b2fb6c33bf1d04ff47dab50a350b1952949","source":{"kind":"arxiv","id":"2506.11595","version":1},"attestation_state":"computed","paper":{"title":"EasyARC: Evaluating Vision Language Models on True Visual Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aylin Akkus, Mert Unsal","submitted_at":"2025-06-13T09:03:33Z","abstract_excerpt":"Building on recent advances in language-based reasoning models, we explore multimodal reasoning that integrates vision and text. Existing multimodal benchmarks primarily test visual extraction combined with text-based reasoning, lacking true visual reasoning with more complex interactions between vision and language. Inspired by the ARC challenge, we introduce EasyARC, a vision-language benchmark requiring multi-image, multi-step reasoning, and self-correction. EasyARC is procedurally generated, fully verifiable, and scalable, making it ideal for reinforcement learning (RL) pipelines. The gene"},"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":"2506.11595","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-13T09:03:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"522c8aba2f5d1f4dabc120426fe5f6cf807be5cca57b49601274bcdd166333c5","abstract_canon_sha256":"3a430566d6a1efd3a5e05d4ff77c106c63e1150ca237dbb4f1b502a0a44b0ccd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:07.338616Z","signature_b64":"Y7n5eeFF7ydS6CqKHfkxE+wEvheEXHuZ6rYswle0YF0uLQJrtt79uV6CnLEhMNkX1CuxIN+huuDbNlsJc8j2Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a3a3b823536e99b82f7935a9a942b2fb6c33bf1d04ff47dab50a350b1952949","last_reissued_at":"2026-07-05T11:21:07.337935Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:07.337935Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EasyARC: Evaluating Vision Language Models on True Visual Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aylin Akkus, Mert Unsal","submitted_at":"2025-06-13T09:03:33Z","abstract_excerpt":"Building on recent advances in language-based reasoning models, we explore multimodal reasoning that integrates vision and text. Existing multimodal benchmarks primarily test visual extraction combined with text-based reasoning, lacking true visual reasoning with more complex interactions between vision and language. Inspired by the ARC challenge, we introduce EasyARC, a vision-language benchmark requiring multi-image, multi-step reasoning, and self-correction. EasyARC is procedurally generated, fully verifiable, and scalable, making it ideal for reinforcement learning (RL) pipelines. The gene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11595","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/2506.11595/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":"2506.11595","created_at":"2026-07-05T11:21:07.338019+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11595v1","created_at":"2026-07-05T11:21:07.338019+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11595","created_at":"2026-07-05T11:21:07.338019+00:00"},{"alias_kind":"pith_short_12","alias_value":"HI5DXARVG3UZ","created_at":"2026-07-05T11:21:07.338019+00:00"},{"alias_kind":"pith_short_16","alias_value":"HI5DXARVG3UZXAXX","created_at":"2026-07-05T11:21:07.338019+00:00"},{"alias_kind":"pith_short_8","alias_value":"HI5DXARV","created_at":"2026-07-05T11:21:07.338019+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01707","citing_title":"LASER: A Corrective Lens for LVLMs via Visual Attention Preservation and Sink Suppression","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16256","citing_title":"Do Vision-Language Models Truly Perform Vision Reasoning? A Rigorous Study of the Modality Gap","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6","json":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6.json","graph_json":"https://pith.science/api/pith-number/HI5DXARVG3UZXAXXSNNJVFBLF6/graph.json","events_json":"https://pith.science/api/pith-number/HI5DXARVG3UZXAXXSNNJVFBLF6/events.json","paper":"https://pith.science/paper/HI5DXARV"},"agent_actions":{"view_html":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6","download_json":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6.json","view_paper":"https://pith.science/paper/HI5DXARV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11595&json=true","fetch_graph":"https://pith.science/api/pith-number/HI5DXARVG3UZXAXXSNNJVFBLF6/graph.json","fetch_events":"https://pith.science/api/pith-number/HI5DXARVG3UZXAXXSNNJVFBLF6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6/action/storage_attestation","attest_author":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6/action/author_attestation","sign_citation":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6/action/citation_signature","submit_replication":"https://pith.science/pith/HI5DXARVG3UZXAXXSNNJVFBLF6/action/replication_record"}},"created_at":"2026-07-05T11:21:07.338019+00:00","updated_at":"2026-07-05T11:21:07.338019+00:00"}