Recent studies have evaluated creativity, where novelty is an important aspect, of large language models (LLMs) primarily from a semantic perspective, using benchmarks from cognitive science. However, assessing the novelty in scholarly publications, a critical facet of evaluating LLMs as scientific discovery assistants, remains underexplored, despite its potential to accelerate research cycles and prioritize high-impact contributions in scientific workflows. We introduce SchNovel, a benchmark to evaluate LLMs’ ability to assess novelty in scholarly papers, a task central to streamlining discovery pipeline. SchNovel consists of 15000 pairs of papers across six fields sampled from the arXiv dataset with publication dates spanning 2 to 10 years apart. In each pair, the more recently published paper is assumed to be more novel. Additionally, we propose RAG-Novelty, a retrieval-augmented method that mirrors human peer review by grounding novelty assessment in retrieved context. Extensive experiments provide insights into the capabilities of different LLMs to assess novelty and demonstrate that RAG-Novelty outperforms recent baseline models highlight LLMs’ promise as tools for automating novelty detection in scientific workflows.

Augmented Language Models (ALMs) empower large language models with the ability to use tools, transforming them into intelligent agents for real-world interactions. However, most existing frameworks for ALMs, to varying degrees, are deficient in the following critical features: flexible customization, collaborative democratization, and holistic evaluation. This paper proposes Gentopia, a lightweight and extensible framework for ALMs. Gentopia allows the flexible customization of agents through simple configurations, seamlessly integrating various language models, task formats, prompting modules, and plugins into a unified paradigm. Furthermore, we establish Gentpool, a public platform enabling the registration and sharing of user-customized agents. Agents registered in Gentpool are composable such that they can be assembled together for agent collaboration, advancing the democratization of artificial intelligence. To ensure high-quality agents, Gentbench, an integral component of Gentpool, is designed to thoroughly evaluate user-customized agents across diverse aspects such as safety, robustness, efficiency, etc. We release Gentopia on Github and will continuously move forward.

Natural language programming automatically generates code based on a user’s text query. Recent solutions are either data-driven or natural language understanding (NLU)-driven. However, the data-driven synthesizer requires a large number of query-code pairs for training, which hinders its application to low-resource programming languages with growing domains whose functionality and grammar can be actively updated. NLU-driven synthesizers solve this problem, but their code generation is slow and their performance rapidly saturates in the presence of ever-increasing data. In this paper, we propose a circular training framework, Colead, which co-evolves both the data-driven synthesizer and the NLU-driven synthesizer to achieve high-quality code generation in the presence of data scarcity and domain growth. The NLU-driven synthesizer generates query-code pairs to update the data-driven synthesizer, which shares a part of its updated model to improve the NLU-driven synthesizers, enabling the co-evolution of both. Experiments show that Colead gives better results than the baselines in the presence of domain growth and data scarcity, and Colead consistently improves the performance of both data-driven and NLU-driven synthesizers over the co-evolvement.