{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.11562", "kind": "paper", "url": "https://arxiv.org/abs/2606.11562"}, {"id": "gh:graphinfer/graphinfer-bench", "kind": "repo", "url": "https://github.com/graphinfer/GraphInfer-Bench"}, {"id": "hf:graphinfer/graphinfer", "kind": "dataset", "url": "https://huggingface.co/datasets/graphinfer/graphinfer"}], "authors": ["Zhuoyi Peng", "Jingzhou Jiang", "Hanlin Gu", "Lixin Fan", "Yi Yang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.11562", "source_url": "https://arxiv.org/abs/2606.11562"}, "categories": ["General AI", "Language & Knowledge", "cs.LG"], "description": {"en": "GraphInfer-Bench evaluates LLMs on graph inference tasks where answers reside in no single node or path, covering five task types over six real-world graphs with 42,000 samples. Tasks include masked-node prediction, edge inference, theme description, outlier detection, and community partition."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.11562", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.11562", "title": "GraphInfer-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Hong Kong University of Science and Technology"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.11562", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.11562", "source_url": "https://openalex.org/W7164396167"}, "key": "claire-radar:2606.11562", "modality": null, "name": "GraphInfer-Bench", "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_graphinfer-bench_ae7fff34", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "77a958b610bbba718c7bf7b7dedda5f2dda393e367ccebffb6c47cb7204177b3", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.11562"}, "publisher": null, "released": "2026-06-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.11562", "source_url": "https://arxiv.org/abs/2606.11562"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-11562", "source": "claire_radar", "source_benchmark_id": "2606.11562", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/graphinfer/GraphInfer-Bench", "githubScope": "benchmark_repo", "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "graphinfer/graphinfer", "hfDatasetDownloads": 74, "hfDatasetLikes": 0, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.11562", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-24", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-24", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "stale"}, "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": "29792f151b116d3a20164d719648aac9e29b1499b1b1a65d41dbeef11782020c", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "GraphInfer-Bench evaluates LLMs on graph inference tasks where answers reside in no single node or path, covering five task types over six real-world graphs with 42,000 samples. Tasks include masked-node prediction, edge inference, theme description, outlier detection, and community partition.", "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 GraphInfer-Bench, a benchmark for whether LLMs can perform this graph inference: producing an open-ended answer that no single node supports and no path retrieves."}, "familyId": "bmf_be514e0e285e", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 1, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "candidates"}, "heat": null, "id": "bm_graphinfer-bench_ae7fff34", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/graphinfer/GraphInfer-Bench", "data": "https://huggingface.co/datasets/graphinfer/graphinfer", "hfPaper": "https://huggingface.co/papers/2606.11562", "paper": "https://arxiv.org/abs/2606.11562", "pdf": "https://arxiv.org/pdf/2606.11562", "project": null, "report": "https://arxiv.org/abs/2606.11562"}, "metrics": [], "motivation": "Graph analysis underlies many applications whose answers cannot be looked up in a single record or retrieved along a path: laundering rings, drug repurposing, user preference, and scientific theme are all inferred from a node together with its neighbourhood.", "name": "GraphInfer-Bench", "oneLine": "GraphInfer-Bench evaluates LLMs on graph inference tasks where answers reside in no single node or path, covering five task types over six real-world graphs with 42,000 samples. Tasks include masked-node prediction, edge inference, theme description, outlier detection, and community partition.", "paperTitle": "GraphInfer-Bench: Benchmarking LLM's Inference Capability on Graphs", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.11562", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "GraphInfer-Bench team", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/graphinfer/GraphInfer-Bench"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-10", "source": {"authors": ["Zhuoyi Peng", "Jingzhou Jiang", "Hanlin Gu", "Lixin Fan", "Yi Yang"], "categories": ["cs.LG", "cs.CL"], "id": "2606.11562", "title": "GraphInfer-Bench: Benchmarking LLM's Inference Capability on Graphs", "type": "arxiv", "url": "https://arxiv.org/abs/2606.11562"}, "sourceUpdatedAt": "2026-06-10T01:41:53Z", "topics": ["cs.LG"], "venueAttempts": [], "whyItMatters": "Targets an open capability gap in graph understanding: inference over joint neighborhood structure. Useful for diagnosing weaknesses in LLMs and GNNs for tasks like fraud detection, drug repurposing, and recommendation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
