Papers

Filtered to scientific domains · clear filter

Browse by term

continual learning 83reinforcement learning 50large language models 12benchmarking 11benchmarks 11language models 11vision-language models 11robotics 7natural language processing 6world models 6generative models 5recursive self-improvement 5attention mechanisms 4diffusion Transformers 4multi-agent systems 4multimodal learning 4multimodal models 4on-policy distillation 4self-distillation 4self-supervised learning 4transformers 4video generation 4vision-language-action models 4agent-based systems 3agentic models 3agentic search 3autonomous systems 3coding agents 3diffusion models 3image generation 3

Matching papers

SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?

61 upvotes · 20 AUG 2026 · Zhipeng Xu, Jiahao Lu, Yining Zheng et al.

This paper introduces SWE-bench Science, a benchmark for evaluating coding agents' performance in repairing scientific software, and identifies common failure mechanisms that hinder their success. Practitioners may care about this research as it can help improve the reliability and reproducibility of scientific findings by developing more effective coding agents.

scientific software engineeringcoding agentsscientific domainsbenchmarkingcontinual learning