Researchers Nicolas Boizard and colleagues investigate whether distilling reasoning traces from strong teacher models is worth the computational cost....
Researchers have proposed a new framework called StruProKGR for efficiently reasoning over sparse knowledge graphs. These graphs are commonly used in ...
Researchers have developed MedRAGChecker, a tool to verify the accuracy of claims made in biomedical text generated by language models. The tool works...
Researchers have developed SlidesGen-Bench, a benchmark for evaluating automated slide generation systems. The benchmark uses computational and quanti...
Researchers have developed a method to improve the accuracy of AI-generated diagrams by incorporating structure and context from the source text. The ...
Researchers propose a three-layer framework for building trustworthy AI systems that support mental health. The framework separates human-oriented tru...
Researchers argue that counterfactual prompting, a method used to evaluate language model bias and faithfulness, is flawed because it doesn't account ...
Researchers have developed a new measure called Granuscore to analyze the level of detail in text. Unlike existing measures that focus on surface-leve...
Researchers at Bilibili have developed a series of open small language models called Index-1.9B. The four models in the series - Index-1.9B-Base, Inde...
Researchers have developed a benchmark called XIH-Bench to evaluate how well language models follow instruction hierarchy rules. This is important bec...
Researchers have developed ZenGen, an integrated framework for measuring and improving social intelligence in large language models (LLMs). The framew...
Researchers have found that coding agents' efficiency cannot be measured solely by token count or model price. They conducted experiments to study the...