{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CFQFRK52VBWYLG7H4DTTAZUIKV","short_pith_number":"pith:CFQFRK52","schema_version":"1.0","canonical_sha256":"116058abbaa86d859be7e0e7306688555f994a01499f80a073bbfdb358b38729","source":{"kind":"arxiv","id":"2505.24876","version":2},"attestation_state":"computed","paper":{"title":"Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Amal Saqib, Fahad Shahbaz Khan, Hanan Ghani, Hisham Cholakkal, Jean Lahoud, Mubarak Shah, Muhra AlMahri, Noor Ahsan, Philip Torr, Rao Muhammad Anwer, Salman Khan, Tajamul Ashraf, Umair Nawaz, Yuhao Li","submitted_at":"2025-05-30T17:59:53Z","abstract_excerpt":"Deep reasoning is fundamental for solving complex tasks, especially in vision-centric scenarios that demand sequential, multimodal understanding. However, existing benchmarks typically evaluate agents with fully synthetic, single-turn queries, limited visual modalities, and lack a framework to assess reasoning quality over multiple steps as required in real-world settings. To address this, we introduce Agent-X, a large-scale benchmark for evaluating vision-centric agents multi-step and deep reasoning capabilities in real-world, multimodal settings. Agent- X features 828 agentic tasks with auth"},"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":"2505.24876","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-30T17:59:53Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"aef7604de0b4286d629306ef6d76ebd3af84e25053521b00ac2f013e647ef58a","abstract_canon_sha256":"8f2c90f1903ee82d33a78465159021d0be1d74047615e375b53c71073811a69c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:12.951143Z","signature_b64":"cGQr2/eygD3nH/LGG2tYwXRz5B3orlkxDOF5Wxia75E928qJsJXFvlfyIDKtZw7TZFC06okWxEwlv6hUWbZfCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"116058abbaa86d859be7e0e7306688555f994a01499f80a073bbfdb358b38729","last_reissued_at":"2026-05-26T01:03:12.950124Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:12.950124Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Amal Saqib, Fahad Shahbaz Khan, Hanan Ghani, Hisham Cholakkal, Jean Lahoud, Mubarak Shah, Muhra AlMahri, Noor Ahsan, Philip Torr, Rao Muhammad Anwer, Salman Khan, Tajamul Ashraf, Umair Nawaz, Yuhao Li","submitted_at":"2025-05-30T17:59:53Z","abstract_excerpt":"Deep reasoning is fundamental for solving complex tasks, especially in vision-centric scenarios that demand sequential, multimodal understanding. However, existing benchmarks typically evaluate agents with fully synthetic, single-turn queries, limited visual modalities, and lack a framework to assess reasoning quality over multiple steps as required in real-world settings. To address this, we introduce Agent-X, a large-scale benchmark for evaluating vision-centric agents multi-step and deep reasoning capabilities in real-world, multimodal settings. Agent- X features 828 agentic tasks with auth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24876","kind":"arxiv","version":2},"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/2505.24876/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":"2505.24876","created_at":"2026-05-26T01:03:12.950477+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24876v2","created_at":"2026-05-26T01:03:12.950477+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24876","created_at":"2026-05-26T01:03:12.950477+00:00"},{"alias_kind":"pith_short_12","alias_value":"CFQFRK52VBWY","created_at":"2026-05-26T01:03:12.950477+00:00"},{"alias_kind":"pith_short_16","alias_value":"CFQFRK52VBWYLG7H","created_at":"2026-05-26T01:03:12.950477+00:00"},{"alias_kind":"pith_short_8","alias_value":"CFQFRK52","created_at":"2026-05-26T01:03:12.950477+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.27955","citing_title":"GUI Agents with Reinforcement Learning: Toward Digital Inhabitants","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV","json":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV.json","graph_json":"https://pith.science/api/pith-number/CFQFRK52VBWYLG7H4DTTAZUIKV/graph.json","events_json":"https://pith.science/api/pith-number/CFQFRK52VBWYLG7H4DTTAZUIKV/events.json","paper":"https://pith.science/paper/CFQFRK52"},"agent_actions":{"view_html":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV","download_json":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV.json","view_paper":"https://pith.science/paper/CFQFRK52","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24876&json=true","fetch_graph":"https://pith.science/api/pith-number/CFQFRK52VBWYLG7H4DTTAZUIKV/graph.json","fetch_events":"https://pith.science/api/pith-number/CFQFRK52VBWYLG7H4DTTAZUIKV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV/action/storage_attestation","attest_author":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV/action/author_attestation","sign_citation":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV/action/citation_signature","submit_replication":"https://pith.science/pith/CFQFRK52VBWYLG7H4DTTAZUIKV/action/replication_record"}},"created_at":"2026-05-26T01:03:12.950477+00:00","updated_at":"2026-05-26T01:03:12.950477+00:00"}