Lu Sun, Xiangyi Zhang, Xiangyang Zhu, Kaiyuan Ji, Zijian Chen, Qiang Hu, Zicheng Zhang, Yuan Tian, Guangtao Zhai · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.26787 · cited by 0 works (OpenAlex) · open-access PDF
Traditionelle Bewertungsbenchmarks reduzieren das inhärent miteinander verbundene wissenschaftliche Wissen in den Lebenswissenschaften in flache Fragenlisten, ohne die zugrunde liegende topologische Struktur des Wissens zu berücksichtigen. Wir stellen den ersten grafisch strukturierten Benchmark für die Lebenswissenschaften ein, der über 10.000 qualitativ hochwertige Multi-Choice-Fragen in Medizin, Biologie und Chemie enthält. Unser Ansatz konstruiert ein gewichtetes Bewertungsdiagramm mithilfe von bidirektionalen Übereinstimmungs- und semantischen Ähnlichkeitsalgorithmen, bei denen Knoten Fragen repräsentieren und Randgewichte ihre semantischen Beziehungen erfassen. Mit dieser grafischen Topologie entwerfen wir zwei neue Bewertungsmetriken.
Machine translation (Meta M2M100); the English original follows.
Traditional evaluation benchmarks reduce inherently interconnected scientific knowledge in life sciences into flat lists of questions, disregarding the underlying topological structure of the knowledge. We introduce, the first graph-structured benchmark for life sciences, featuring over 10,000 high-quality multiple-choice questions across medicine, biology, and chemistry. Our approach constructs a weighted evaluation graph using bidirectional matching and semantic similarity algorithms, where nodes represent questions and edge weights capture their semantic relationships. Leveraging this graph topology, we design two novel evaluation metrics. The Global Coherence Score (GCS) measures a model's consistency within semantically related neighborhoods, while Knowledge Balance Score (KBS) analyzes how model errors are distributed across the graph to reveal conceptual blind spots. LG-Bench facilitates fine-grained comparison of LLMs by surfacing differences in conceptual coherence and patterns of knowledge organization across models. Our framework shifts the evaluation paradigm from flat accuracy metrics to structure-aware analysis, offering a new lens for diagnosing and improving LLM performance in the life sciences domain.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Computer science, Benchmark (surveying), Consistency (knowledge bases), Graph, Coherence (philosophical gambling strategy), Matching (statistics)
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26787, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange