{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.10408", "kind": "paper", "url": "https://arxiv.org/abs/2608.10408"}, {"id": "gh:vis-nlp/viseditbench", "kind": "repo", "url": "https://github.com/vis-nlp/VisEditBench"}], "authors": ["Mizanur Rahman", "Arshia Azimlu", "Shadikur Rahman", "Md Tahmid Rahman Laskar", "Amran Bhuiyan", "Shafiq Joty", "Enamul Hoque Prince"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.10408", "source_url": "https://arxiv.org/abs/2608.10408"}, "categories": ["General AI", "Multimodal"], "description": {"en": "VisEditBench is a benchmark comprising 1,395 human-annotated visualization code-editing tasks across two settings: feedback-guided repair and reference-guided restyling. Models are evaluated on their ability to revise existing visualization code based on multimodal feedback such as buggy or marked charts with textual instructions, and target chart images. Scoring is based on pass rates of generated code executions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.10408", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.10408", "title": "VisEditBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.10408", "modality": null, "name": "VisEditBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_viseditbench_e7b0411b", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "47bd3adbb18639690268d68b3f30162e8752c7ad8b42e343875ebd9695c7b37a", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.10408"}, "publisher": null, "released": "2026-08-11", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.10408", "source_url": "https://arxiv.org/abs/2608.10408"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-10408", "source": "claire_radar", "source_benchmark_id": "2608.10408", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/vis-nlp/VisEditBench", "githubScope": "benchmark_repo", "githubStars": 3, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.10408", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "0140f504284b616dcae006538da4045f8c23ddf98da04ed0a9a58224967c4c7e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "VisEditBench is a benchmark comprising 1,395 human-annotated visualization code-editing tasks across two settings: feedback-guided repair and reference-guided restyling. Models are evaluated on their ability to revise existing visualization code based on multimodal feedback such as buggy or marked charts with textual instructions, and target chart images. Scoring is based on pass rates of generated code executions.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce VisEditBench, a benchmark of 1,395 human-annotated visualization code-editing tasks grounded in realistic visualization workflows and failure cases."}, "familyId": "bmf_e2e3088b5cf4", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/vis-nlp/VisEditBench", "url": "https://github.com/vis-nlp/VisEditBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/vis-nlp/VisEditBench", "url": "https://huggingface.co/papers/2608.10408"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/vis-nlp/VisEditBench", "url": "https://arxiv.org/abs/2608.10408"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/vis-nlp/VisEditBench.", "url": "https://arxiv.org/abs/2608.10408"}], "scope": "benchmark_repo", "url": "https://github.com/vis-nlp/VisEditBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_viseditbench_e7b0411b", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/vis-nlp/VisEditBench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.10408", "paper": "https://arxiv.org/abs/2608.10408", "pdf": "https://arxiv.org/pdf/2608.10408", "project": null, "report": "https://arxiv.org/abs/2608.10408"}, "metrics": [], "motivation": "Vision-language models (VLMs) have shown strong capabilities in generating visualization code from textual or visual specifications.", "name": "VisEditBench", "oneLine": "VisEditBench is a benchmark comprising 1,395 human-annotated visualization code-editing tasks across two settings: feedback-guided repair and reference-guided restyling. Models are evaluated on their ability to revise existing visualization code based on multimodal feedback such as buggy or marked charts with textual instructions, and target chart images. Scoring is based on pass rates of generated code executions.", "paperTitle": "VisEditBench: Can Vision-Language Models Edit Visualization Code from Multimodal Feedback?", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.10408", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "vis-nlp", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/vis-nlp/VisEditBench"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.55, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.18504537701418716, "value": 3}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 226, "score": 33}, "method": "attention-ranking-v13", "rank": 226, "score": 33}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-11", "source": {"authors": ["Mizanur Rahman", "Arshia Azimlu", "Shadikur Rahman", "Md Tahmid Rahman Laskar", "Amran Bhuiyan", "Shafiq Joty", "Enamul Hoque Prince"], "categories": ["cs.CL"], "id": "2608.10408", "title": "VisEditBench: Can Vision-Language Models Edit Visualization Code from Multimodal Feedback?", "type": "arxiv", "url": "https://arxiv.org/abs/2608.10408"}, "sourceUpdatedAt": "2026-08-11T02:52:23Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Existing benchmarks focus on generating visualizations from scratch, leaving the iterative editing process unexplored. VisEditBench provides a standardized evaluation for this practical task, enabling comparisons across VLMs and highlighting gaps in open-source models, particularly in visually grounded style adaptation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
