{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OU5Q6JBCNNHHG6LMFKX3LAG45J","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":"5c2192600db12b1070f8f5e9612cf07396e4e555763c9704013b3dd30493def5","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T05:39:38Z","title_canon_sha256":"9c917cbe3ab626a3c6082f975b679434e6e71edd7eca2c6dd2e85f1627553f49"},"schema_version":"1.0","source":{"id":"2605.13034","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13034","created_at":"2026-05-18T03:08:59Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13034v1","created_at":"2026-05-18T03:08:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13034","created_at":"2026-05-18T03:08:59Z"},{"alias_kind":"pith_short_12","alias_value":"OU5Q6JBCNNHH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"OU5Q6JBCNNHHG6LM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"OU5Q6JBC","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:9d2d6272f4387def7bcd2dfc80d5190db0bd3b87213ddd15e4f435503a95defa","target":"graph","created_at":"2026-05-18T03:08:59Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experiments show that ViDR improves overall report quality, source-figure integration, and verifiability over strong commercial and open-source baselines."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That context-aware filtering, outline-aware reranking, and VLM-based visual analysis can reliably turn noisy web images into accurate, non-hallucinated evidence atoms without introducing new errors that affect report claims."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ViDR treats source figures as retrievable and verifiable evidence objects in multimodal deep research reports and introduces MMR Bench+ to measure improvements in visual integration and verifiability."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Treating source figures as verifiable evidence objects improves the quality and verifiability of multimodal deep research reports."}],"snapshot_sha256":"37c97e4ec82ec3d1c004641d8a493fa9dd5388ae63af924e73133a00b96039b6"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"4031afe3658214c0f334248249b98857643e87468042d0595f794506fadbb6cf"},"paper":{"abstract_excerpt":"Recent deep research systems have improved the ability of large language models to produce long, grounded reports through iterative retrieval and reasoning. However, most text-centered systems rely mainly on textual evidence, while multimodal systems often retrieve images only weakly or generate charts themselves, leaving source figures underused as evidence. We present ViDR, a multimodal deep research framework that grounds long-form reports in source figures. ViDR treats source figures as retrievable, interpretable, routable, and verifiable evidence objects, while still generating analytical","authors_text":"Baoqin Sun, Haiyang Shen, Peilun Jia, Sixiong Xie, Xiang Jing, Yun Ma, Zhuofan Shi","cross_cats":["cs.IR"],"headline":"Treating source figures as verifiable evidence objects improves the quality and verifiability of multimodal deep research reports.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T05:39:38Z","title":"ViDR: Grounding Multimodal Deep Research Reports in Source Visual Evidence"},"references":{"count":36,"internal_anchors":7,"resolved_work":36,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Try deep research and our new experimental model in gemini, your ai assistant","work_id":"942e9dd3-f690-4939-a3d1-2163373696f8","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Paladugu, Pranav Setlur, Jiahe Jin, James P","work_id":"2f6b8cd2-51ef-4670-8361-26e737e919cf","year":2025},{"cited_arxiv_id":"2506.11763","doi":"","is_internal_anchor":true,"ref_index":3,"title":"DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents","work_id":"449edaa7-8f9d-4170-952d-4ab0d740572c","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"gpt-researcher","work_id":"537bce4c-bf4b-4dbb-a350-0a4566ccb2c4","year":2025},{"cited_arxiv_id":"2508.05748","doi":"","is_internal_anchor":true,"ref_index":5,"title":"WebWatcher: Breaking New Frontier of Vision-Language Deep Research Agent","work_id":"b734fd16-d7cf-4129-ac4d-129f31964890","year":2025}],"snapshot_sha256":"7b180e11774dfcdd6a3c082a5079c29725bc30831ab5e214611568ee440162ed"},"source":{"id":"2605.13034","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T19:15:11.787518Z","id":"7b067368-07aa-4317-8a46-d0b3f01a293c","model_set":{"reader":"grok-4.3"},"one_line_summary":"ViDR treats source figures as retrievable and verifiable evidence objects in multimodal deep research reports and introduces MMR Bench+ to measure improvements in visual integration and verifiability.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Treating source figures as verifiable evidence objects improves the quality and verifiability of multimodal deep research reports.","strongest_claim":"Experiments show that ViDR improves overall report quality, source-figure integration, and verifiability over strong commercial and open-source baselines.","weakest_assumption":"That context-aware filtering, outline-aware reranking, and VLM-based visual analysis can reliably turn noisy web images into accurate, non-hallucinated evidence atoms without introducing new errors that affect report claims."}},"verdict_id":"7b067368-07aa-4317-8a46-d0b3f01a293c"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c191d1061235af16bebfed25204fa0e4d6d818f2a1505ccc27db8894d53bf1b7","target":"record","created_at":"2026-05-18T03:08:59Z","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":"5c2192600db12b1070f8f5e9612cf07396e4e555763c9704013b3dd30493def5","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T05:39:38Z","title_canon_sha256":"9c917cbe3ab626a3c6082f975b679434e6e71edd7eca2c6dd2e85f1627553f49"},"schema_version":"1.0","source":{"id":"2605.13034","kind":"arxiv","version":1}},"canonical_sha256":"753b0f24226b4e73796c2aafb580dcea5ec446ce6a16d81fdd4bc30892f79c64","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"753b0f24226b4e73796c2aafb580dcea5ec446ce6a16d81fdd4bc30892f79c64","first_computed_at":"2026-05-18T03:08:59.668216Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:08:59.668216Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3v8Qxvet45yKLZl5HYss+ksQ0IFHls+DJiNw3JUpjV2+Gp25L1fxDAPCjcZsNeYHr+sWQxE0JVYGSSXa/UE6Cg==","signature_status":"signed_v1","signed_at":"2026-05-18T03:08:59.668781Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13034","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c191d1061235af16bebfed25204fa0e4d6d818f2a1505ccc27db8894d53bf1b7","sha256:9d2d6272f4387def7bcd2dfc80d5190db0bd3b87213ddd15e4f435503a95defa"],"state_sha256":"fb4eb15df17cdd2dfa63602422cbb75b7bbcbe2e6c6d83305a7b8263b9d8ceec"}