Papers

Filtered to shadow evaluations · clear filter

Browse by term

continual learning 64reinforcement learning 34large language models 17benchmarking 12vision-language models 10generative models 8language models 8video generation 7multimodal models 6natural language processing 6robotics 6world models 6benchmarks 5diffusion models 5on-policy distillation 5policy optimization 5scalability 5self-distillation 5vision-language-action models 5autoregressive models 4computer vision 4diffusion transformers 4LLMs 4multimodal large language models 4verifiable rewards 4attention mechanisms 3embodied intelligence 3image editing 3long-term memory 3multimodal learning 3

Matching papers

Can AI agents conduct open-ended AI research? Early evidence from two case studies

11 upvotes · 29 JUL 2026 · Peter Kirgis, Sayash Kapoor, Andrew Schwartz et al.

This paper investigates whether AI agents can conduct open-ended AI research and provides early evidence that they can perform the engineering aspects but struggle with critical parts of the research lifecycle, such as making progress on research questions and judgment about publishable research.

AI R\&Dopen-ended researchshadow evaluationscontinual learningagent performancebenchmarking