Tag
This paper introduces DoLQ, a multi-agent framework that uses Large Language Models to perform both qualitative and quantitative evaluations for discovering ordinary differential equations from observational data.
This paper introduces MOSAIC, a method for module discovery in scientific time series that combines causal representation learning with sparse additive identifiable causal learning. It aims to recover interpretable latent variables and their associated observations without post-hoc alignment, validated on domains like molecular dynamics and climate data.
DeepMind highlights the expanded impact of AlphaEvolve, a Gemini-powered coding agent, demonstrating its ability to optimize algorithms for genomics, grid optimization, earth sciences, quantum physics, and mathematics.
Columbia CS Prof Vishal Misra argues LLMs can’t generate truly novel science because they only interpolate within learned Bayesian manifolds rather than create new conceptual maps.
SimpleTES framework scales evaluation-driven discovery loops across 21 scientific problems, yielding 2× speedups on LASSO, 24.5% quantum gate reductions, and new Erdos constructions while enabling trajectory-level model post-training.
EvoMaster is a scalable, self-evolving agent framework for large-scale scientific discovery that enables iterative hypothesis refinement and knowledge accumulation across experimental cycles. It achieves state-of-the-art results on four benchmarks including Humanity's Last Exam (41.1%) and MLE-Bench Lite (75.8%), outperforming general-purpose baselines by up to 316%.
PRL-Bench is a comprehensive benchmark for evaluating LLMs' capabilities in frontier physics research, constructed from 100 curated Physical Review Letters papers across five physics subfields. The benchmark reveals significant gaps in current LLM performance (best scores below 50%), designed to test end-to-end research workflows, complex reasoning, and autonomous exploration.
EvoScientist is an adaptive multi-agent framework for end-to-end scientific discovery that continuously improves through persistent memory modules, comprising three specialized agents for idea generation, experiment execution, and knowledge distillation. It outperforms 7 state-of-the-art systems in scientific idea generation and improves code execution success rates through multi-agent evolution.
Google reviews its 2025 research breakthroughs, highlighting advancements in Gemini 3 and Gemma 3 models, AI agents, and scientific applications across computing and healthcare.
OpenAI releases a research paper documenting early experiments where GPT-5 accelerates scientific discovery across mathematics, physics, biology, and other fields, including cases where the model helped identify disease mechanisms, solve open problems, and improve optimization algorithms in collaboration with leading universities and national laboratories.
Google DeepMind and Google launched the AI for Math Initiative, partnering with five prestigious research institutions to use AI tools like Gemini Deep Think and AlphaProof to accelerate mathematical discoveries.
DeepMind researchers discovered new families of unstable singularities in fundamental fluid dynamics equations using AI techniques, potentially advancing understanding of century-old mathematical problems like the Navier-Stokes equations. The work collaborates with Brown, NYU, and Stanford, revealing patterns in blow-up behavior with unprecedented computational accuracy.
OpenAI organized a "1,000 Scientist AI Jam Session" across nine U.S. Department of Energy national labs, bringing together over 1,000 scientists to test advanced AI models like o3-mini for accelerating scientific discovery. The event demonstrates a major public-private collaboration aimed at strengthening U.S. AI leadership while addressing research challenges in materials science, renewable energy, and astrophysics.
At the I/O 2026 keynote, Google announced the latest AI achievements including Gemini 3.5, Omni, Code Mender, Gemini for Science, WeatherNext, and emphasized that AGI is imminent, and demonstrated AI applications in cybersecurity, scientific simulation, and health.