Large language models perform well on many logical reasoning benchmarks, but it remains unclear which core logical skills they truly master. To address this, we introduce LogicSkills, a benchmark that isolates three fundamental logical skills: (i) $\textit{formal symbolization}\unicode{x2014}{}$translating premises into first-order logic; (ii) $\textit{countermodel construction}\unicode{x2014}$showing that an argument is logically invalid by constructing a finite countermodel; and (iii) $\textit{validity assessment}\unicode{x2014}$determining whether a conclusion follows from a set of premises. Items are drawn from the two-variable fragment of first-order logic without identity and are presented in both English and a Carrollian nonce-word language. All instances are solver-verified with Z3 for correctness and non-triviality. Across conventional instruction-tuned LLMs, performance is high on $\textit{validity assessment}$ but substantially lower on $\textit{formal symbolization}$ and $\textit{countermodel construction}$, highlighting that high task-level accuracy can mask weaknesses in core logical skills. In contrast, recent reasoning-tuned models perform strongly across all three tasks, suggesting a more systematic logical skill profile.
OneEval: Benchmarking LLM Knowledge-intensive Reasoning over Diverse Knowledge Bases
Southeast University（东南大学）；Zhejiang University（浙江大学）；Nanjing University of Posts and Telecommunications（南京邮电大学）；Tongji University（同济大学）；Monash University（莫纳什大学）；Soochow University（苏州大学）
Large Language Models for Structured Clinical Data Analysis: Dual-Agent Grounding and Validation
Harvard Medical School（哈佛医学院）；Wayne State University School of Medicine（韦恩州立大学医学院）；Women & Infants Hospital of Rhode Island（罗德岛妇女与婴儿医院）；Warren Alpert Medical School of Brown University（布朗大学沃伦·阿尔珀特医学院）；Duke University School of Medicine（杜克大学医学院）
VeryTrace: Verifying Reasoning Traces through Compilable Formalism and Structured Verification
School of Electrical and Computer Engineering, Georgia Institute of Technology, USA（佐治亚理工学院电子与计算机工程学院）；Department of Electrical and Computer Engineering, Bogazici University, Turkey（博亚奇大学电子与计算机工程系）
我们扩展了结构化论证的$ASPIC^+$框架，以便允许应用的情况下，推理方案的可废止的参数。给定一个带有结论“$A$ or $B$”的论证，一个基于$A$带有结论$C$的论证，以及一个基于$B$带有结论$C$的论证，我们允许构造一个带有结论$C$的论证。我们展示了我们的框架如何导致不同的结果比其他方法在非单调逻辑处理析取信息，如析取默认理论或方法的基础上的OR规则（允许导出一个可废止的规则“如果（$A$或$B$），然后$C$”，给出两个可废止的规则“如果$A$，然后$C$”和“如果$B$，然后$C$”）。我们提出了新的问题，推理的微妙之处defeably与析取信息，并表明其形式化是更复杂的比一个人会假设。
1st ICLR International Workshop on Privacy, Accountability, Interpretability, Robustness, Reasoning on Structured Data (PAIR^2Struct)
Rutgers University（罗格斯大学）；Hong Kong Polytechnic University（香港理工大学）；Massachusetts Institute of Technology（麻省理工学院）；King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；Columbia University（哥伦比亚大学）；Peking University（北京大学）
