Issues

• 对于标准答案的疑惑

  GraphRAG-Bench/GraphRAG-Benchmark#62 · Filter by author xingzhao-tt opened on Oct 14, 2026

• generation_eval: several failure modes silently produce 0 / NaN scores or drop samples (non-OpenAI judges, long answers)

  GraphRAG-Bench/GraphRAG-Benchmark#61 · Filter by author bhanuswami026 opened on Oct 10, 2026

• EvalPort: portable interchange for the per-sample eval record in Evaluation/retrieval_eval.py

  GraphRAG-Bench/GraphRAG-Benchmark#60 · Filter by author adhabnr-ux opened on Sep 27, 2026

• Question: Handling chunks longer than BGE-large-en-v1.5’s 512-token limit

  GraphRAG-Bench/GraphRAG-Benchmark#58 · Filter by author lixiaoguai11 opened on Jul 30, 2026

• 关于复现指标差异与建图报错问题

  GraphRAG-Bench/GraphRAG-Benchmark#57 · Filter by author Lx-zjwf opened on Jun 18, 2026

• Measuring answer generation this way is biased

  GraphRAG-Bench/GraphRAG-Benchmark#55 · Filter by author fIa5h opened on Apr 29, 2026
  1

• 问题分类分布咨询

  GraphRAG-Bench/GraphRAG-Benchmark#52 · Filter by author cskkxjk opened on Apr 8, 2026
  1