{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LTWHYP5BK3AXMP4XSVVRIRNI6Q","short_pith_number":"pith:LTWHYP5B","schema_version":"1.0","canonical_sha256":"5cec7c3fa156c1763f97956b1445a8f42f43b136365004d76ce88b6d7247b0c5","source":{"kind":"arxiv","id":"2605.10832","version":2},"attestation_state":"computed","paper":{"title":"Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"On-policy data evolution from agent rollouts boosts multimodal deep search performance from 24.9% to 39% on average.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenxin Li, Guanting Dong, Hangyu Guo, Hongru Wang, Junting Lu, Shijue Huang, Shuang Chen, Xinyu Geng, Yi R. Fung, Zhaochen Su, Zhenyu Li","submitted_at":"2026-05-11T16:49:36Z","abstract_excerpt":"Multimodal deep search requires an agent to solve open-world problems by chaining search, tool use, and visual reasoning over evolving textual and visual context. Two bottlenecks limit current systems. First, existing tool-use harnesses treat images returned by search, browsing, or transformation as transient outputs, so intermediate visual evidence cannot be re-consumed by later tools. Second, training data is usually built by fixed curation recipes that cannot track the target agent's evolving capability. To address these challenges, we first introduce a visual-native agent harness centered "},"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":"2605.10832","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-11T16:49:36Z","cross_cats_sorted":[],"title_canon_sha256":"e178d81e90584fcf37865ee23101edc46bb6f4f5cdd66e76965a755c80b267b4","abstract_canon_sha256":"f3a288909e3ddee0339012bde8bc17676df114c7ef398a2f4b48d6150606d5d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:04:07.309241Z","signature_b64":"mQjsZL7omd1buuO+1dRQDhpA0Sco6PmjfKeJQLj9WGAAQa8vmb3HMHiE5WBWBc6Y326HAlBD1OzKYRtmWBaIDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cec7c3fa156c1763f97956b1445a8f42f43b136365004d76ce88b6d7247b0c5","last_reissued_at":"2026-06-08T01:04:07.308478Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:04:07.308478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"On-policy data evolution from agent rollouts boosts multimodal deep search performance from 24.9% to 39% on average.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenxin Li, Guanting Dong, Hangyu Guo, Hongru Wang, Junting Lu, Shijue Huang, Shuang Chen, Xinyu Geng, Yi R. Fung, Zhaochen Su, Zhenyu Li","submitted_at":"2026-05-11T16:49:36Z","abstract_excerpt":"Multimodal deep search requires an agent to solve open-world problems by chaining search, tool use, and visual reasoning over evolving textual and visual context. Two bottlenecks limit current systems. First, existing tool-use harnesses treat images returned by search, browsing, or transformation as transient outputs, so intermediate visual evidence cannot be re-consumed by later tools. Second, training data is usually built by fixed curation recipes that cannot track the target agent's evolving capability. To address these challenges, we first introduce a visual-native agent harness centered "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across 8 multimodal deep search benchmarks, ODE improves the Qwen3-VL-8B agent from 24.9% to 39.0% on average, surpassing Gemini-2.5 Pro in standard agent-workflow setting (37.9%). At 30B, ODE raises the average score from 30.6% to 41.5%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That rollouts from the current policy produce training data that accurately identifies and fills the precise capability gaps without introducing self-reinforcing biases or training instability.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A new image-bank harness and closed-loop on-policy data evolution method raises multimodal agent performance on visual search benchmarks from 24.9% to 39.0% for an 8B model and from 30.6% to 41.5% for a 30B model.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"On-policy data evolution from agent rollouts boosts multimodal deep search performance from 24.9% to 39% on average.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5bb6ee0af1ca64bfb9cf6447b9a1bea750562ba4153f266f382f115c32e5b2a0"},"source":{"id":"2605.10832","kind":"arxiv","version":2},"verdict":{"id":"7a3df81e-29a5-422c-ad95-61721d7662e6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T04:10:59.872308Z","strongest_claim":"Across 8 multimodal deep search benchmarks, ODE improves the Qwen3-VL-8B agent from 24.9% to 39.0% on average, surpassing Gemini-2.5 Pro in standard agent-workflow setting (37.9%). At 30B, ODE raises the average score from 30.6% to 41.5%.","one_line_summary":"A new image-bank harness and closed-loop on-policy data evolution method raises multimodal agent performance on visual search benchmarks from 24.9% to 39.0% for an 8B model and from 30.6% to 41.5% for a 30B model.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That rollouts from the current policy produce training data that accurately identifies and fills the precise capability gaps without introducing self-reinforcing biases or training instability.","pith_extraction_headline":"On-policy data evolution from agent rollouts boosts multimodal deep search performance from 24.9% to 39% on average."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.10832/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T05:22:00.342835Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T14:34:47.442439Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T10:31:17.587837Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:57:25.750037Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"d5d3cb52ab2d139c70b8feedfeceeb355eb45c3f7582d09b4f05e0f3292d6e6a"},"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":"2605.10832","created_at":"2026-06-08T01:04:07.308583+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.10832v2","created_at":"2026-06-08T01:04:07.308583+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.10832","created_at":"2026-06-08T01:04:07.308583+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTWHYP5BK3AX","created_at":"2026-06-08T01:04:07.308583+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTWHYP5BK3AXMP4X","created_at":"2026-06-08T01:04:07.308583+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTWHYP5B","created_at":"2026-06-08T01:04:07.308583+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/LTWHYP5BK3AXMP4XSVVRIRNI6Q","json":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q.json","graph_json":"https://pith.science/api/pith-number/LTWHYP5BK3AXMP4XSVVRIRNI6Q/graph.json","events_json":"https://pith.science/api/pith-number/LTWHYP5BK3AXMP4XSVVRIRNI6Q/events.json","paper":"https://pith.science/paper/LTWHYP5B"},"agent_actions":{"view_html":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q","download_json":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q.json","view_paper":"https://pith.science/paper/LTWHYP5B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.10832&json=true","fetch_graph":"https://pith.science/api/pith-number/LTWHYP5BK3AXMP4XSVVRIRNI6Q/graph.json","fetch_events":"https://pith.science/api/pith-number/LTWHYP5BK3AXMP4XSVVRIRNI6Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q/action/storage_attestation","attest_author":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q/action/author_attestation","sign_citation":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q/action/citation_signature","submit_replication":"https://pith.science/pith/LTWHYP5BK3AXMP4XSVVRIRNI6Q/action/replication_record"}},"created_at":"2026-06-08T01:04:07.308583+00:00","updated_at":"2026-06-08T01:04:07.308583+00:00"}