EurekaBench: Measuring Agentic Ability to Discover New Scientific Insights

Zhijing Jin on Twitter / X

🚀 Can #AI agents actually do science?Current agents can optimize for predictive fit — but fail at recovering the underlying physical laws.We introduce Stargazer🪐: a scalable benchmark + environment for astronomical discovery🔭Agents must:• propose hypotheses•… pic.twitter.com/hGIjz4KJMT— Zhijing Jin (@ZhijingJin) May 7, 2026

ScientistTwo: Pioneering the Human Knowledge Frontier with Autonomous AI

Scientific discovery is defined by the ability to identify the boundaries of existing knowledge and venture into unexplored territory. The ultimate vision for AI in science is problem-driven autonomous research: given a fundamental challenge by a human expert, the AI independently navigates the scientific landscape, uncovers theoretical and empirical bottlenecks, and systematically expands the frontier of knowledge. In this paper, we introduce ScientistTwo, a fully autonomous multi-agent framework designed to realize this vision. Specifically, ScientistTwo takes an initial problem as input, establishes state-of-the-art baselines, formulates novel hypotheses, and coordinates specialized agents to orchestrate an end-to-end discovery cycle without human intervention. Moreover, the framework rigorously conducts experiments using diverse datasets and metrics, refines methodologies through automated ablation studies, and validates research findings via a closed-loop simulated peer-review rebuttal engine. To evaluate ScientistTwo's capabilities against the highest standards of human scientific achievement, we benchmark it across papers accepted at top-tier conferences such as ICLR, ICML, and NeurIPS. As a result, ScientistTwo autonomously generates expert-level, publishable papers and fully verified, executable codebases. Its solutions consistently outperform human state-of-the-art models, and achieve higher average review ratings than human-authored papers under automated AI review agents. These results show that ScientistTwo is not merely an assistive tool but an autonomous scientific pioneer capable of pushing the frontiers of human discovery. Project website: https://scientist-two.github.io/

Can AI Agents Synthesize Scientific Conclusions?

Scientific AI agents increasingly retrieve evidence, reason across sources, and synthesize conclusions used in consequential decisions. Yet, their ability to do so in high-stakes domains such as health remains unclear. We introduce SciConBench, a large-scale live benchmark of 9.11K questions and expert-written conclusions from systematic reviews to evaluate open-domain scientific conclusion synthesis. The benchmark draws on an expert-validated automated evaluation pipeline that decomposes conclusions into atomic facts and measures correctness and comprehensiveness via factual precision and recall. To mitigate data leakage, we further introduce SciConHarness, a clean-room evaluation harness that equips agents with controlled web interaction to ensure valid measurement. Evaluating 8 frontier models and deep research agents, we find that factual quality remains low: under clean-room settings, the best agent achieves only a factual F1 of 0.337. Our clean-room setting consistently reduces performance relative to unconstrained evaluation, suggesting that leakage inflates estimates of models' true synthesis capabilities. Finally, we audit consumer-facing agents (e.g., Google AI Overview, OpenEvidence) and find they frequently generate incomplete and sometimes contradictory conclusions, even when the ground-truth answer is available. Overall, our results show that reliable synthesis of scientific conclusions remains an open challenge, and that clean-room evaluation is essential for assessing open-domain AI agents.

Verification in the Age of AI Scientists

AI Scientists now operate at a scale that outpaces human capacity for manual review. Systems such as Sakana’s AI Scientist write entire workshop papers end-to-end. Lila Sciences runs autonomous ``AI Science Factories’’ that hypothesize, experiment, and iterate without human guidance. FutureHouse’s Kosmos and Robin generate thousands of candidate hypotheses in a single run, and Google’s Co-Scientist proposes testable experiments at a rate no laboratory can fully evaluate. Each of these systems emphasizes verified results, yet that standard ranges from near-perfect formal proof in mathematics to decade-long clinical trials in medicine, with no shared framework for judging sufficiency across domains. As outputs scale beyond what humans can manually inspect, the question of which results to trust becomes as hard as generating them. The bottleneck for AI for Science is no longer hypothesis generation, it is verification.

AI Scientists as Engines of Discovery: A Case for Development within Reformed Institutions

Agentic artificial intelligence (AI) systems are beginning to assist, accelerate, and partially automate scientific discovery, performing tasks that span literature synthesis, code generation, data analysis, hypothesis proposal, and model criticism. We argue that this transition is qualitative rather than incremental, and that suitably designed multi-agent systems may evolve from passive computational tools into ``AI scientists'' that can expand the hypothesis-generating and verification capacity of science. Such systems must be developed and deployed within a scientific ecosystem fit for purpose: institutions must be redesigned for verification, accountability, interpretability, and dual-use safety. We sketch how multi-agent architectures, illustrated by the prototype framework \textit{Denario}, accelerate the discovery cycle and traverse model spaces beyond human reach; examine what this implies for authorship, peer review, and the enduring role of human scientists; and close with recommendations for governing AI as an epistemic actor rather than a mere instrument.

Agentic Laboratories of the Future: Towards World Models for Scientific Discovery

Scientific discovery is fundamentally a problem-solving process involving distributed intelligence. Human intuition, computational reasoning, and experimental execution are distributed across people, instruments, and software systems, limiting the speed and scale of discovery. Although automation, high-throughput experimentation, foundation models, and cloud infrastructure have accelerated individual stages of the scientific workflow, they have not unified the discovery process. We hypothesize that the next generation of laboratories will be agentic: environments in which scientists, AI systems, and robotic platforms operate as collaborative discovery partners, with humans contributing the parts of discovery that remain hardest to make explicit: asking the right questions and holding provisional mechanistic models of how a system works. The key missing layer is an agentic harnessing layer that continuously integrates hypothesis, literature-derived evidence, experimental data, uncertainty, and experimental state into a shared “laboratory world model”—a dynamic representation of the scientific system and its evolving context. By maintaining and updating this lab world model, the agentic harnessing layer enables coordinated decision-making, adaptive planning, and increasingly autonomous scientific workflows across humans and machines. A central challenge is that much of the scientific research process remains inaccessible to machines, including tacit knowledge, human observations, adaptive decision-making, and evolving experimental context. Advances in multimodal AI and immersive interfaces may help bridge this gap, allowing humans, agents, and robotic systems to collaborate seamlessly in scientific discovery rather than simply automating isolated tasks. Agentic laboratories could provide a new architecture for science, integrating human, artificial, and physical intelligence into a unified discovery system.

AI-Newton: A Concept-Driven Physical Law Discovery System without Prior Physical Knowledge

While current AI-driven methods excel at deriving empirical models from individual experiments, a significant challenge remains in uncovering the common fundamental physics that underlie these models -- a task at which human physicists are adept. To bridge this gap, we introduce AI-Newton, a novel framework for concept-driven scientific discovery. Our system autonomously derives general physical laws directly from raw, multi-experiment data, operating without supervision or prior physical knowledge. Its core innovations are twofold: (1) proposing interpretable physical concepts to construct laws, and (2) progressively generalizing these laws to broader domains. Applied to a large, noisy dataset of mechanics experiments, AI-Newton successfully rediscovers foundational and universal laws, such as Newton's second law, the conservation of energy, and the universal gravitation. This work represents a significant advance toward autonomous, human-like scientific discovery.

ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments

Scientific code repositories encode decades of human knowledge in executable models, methods, and tools. Yet fragmented toolchains, implicit domain conventions, and specialized correctness criteria make this knowledge difficult to convert into reliable learning experience-a challenge we call the scientific experience bottleneck. We introduce ScienceIDE, infrastructure for turning the world's scientific code into programmable environments for scientific agents. Guided by expert-defined scientific cases and acceptance criteria, agents transform repositories into executable environments that support task generation, execution, and scientific verification. These environments provide a shared foundation for supervised fine-tuning, reinforcement learning, and evaluation. Using verified interaction trajectories, we train PhAI-IDE-72B, PhAI-IDE-9B, and PhAI-IDE-4B. The model family shows gains in held-out scientific-code repair and across selected general-purpose benchmarks in code, reasoning, and knowledge, providing evidence of positive transfer from scientific experience to broader capabilities. ScienceIDE lays the foundation for an integrated workspace for agent learning and scientific practice, making humanity's scientific software a shared substrate for developing scientific intelligence. Code: https://github.com/aitofound/ScienceIDE

Towards Automating Scientific Review with Google's Paper Assistant Tool

Artificial intelligence is driving a revolution in scientific discovery, accelerating everything from hypothesis generation to mathematical theorem proving. However, this rapid acceleration is creating a systemic challenge: traditional human peer review cannot scale to match the influx of AI-assisted science. Ultimately, to resolve this tension, we must also deploy AI to accelerate the verification and review process itself. To frame the discussion around this transition, we propose a taxonomy consisting of four progressive levels of AI-human collaboration in scientific evaluation, and discuss various trade-offs involved with each. As a step toward this future, we introduce the Paper Assistant Tool (PAT), an agentic AI framework built for deep scientific review and verification. PAT ingests full scientific manuscripts and produces a comprehensive evaluation, checking theoretical results, validating experiments, suggesting improvements, and identifying potential flaws. By utilizing inference scaling techniques, PAT is able to identify deeper issues than a single model call alone, achieving a 34% improvement over zero-shot recall on mathematical errors in the SPOT benchmark. Pilot deployments of PAT as a pre-submission tool for authors at two major Computer Science conferences -- STOC and ICML -- demonstrate its ability to identify critical errors and suggest substantive improvements to research papers. By catching errors early, PAT eases the cognitive burden placed on referees, while preserving their control over the outcomes of the review process.

Training AI Scientists to Replicate Research

The replicability of papers is a cornerstone of scientific knowledge, ensuring the reliability of existing results and providing a base for further experiments. The act of replication typically illuminates details that were previously underspecified, and thus requires similar hypothesis-driven exploration to open-ended research. In this work, we develop Replica, a scalable task space for paper replication. To provide reward signal, we introduce an auto-generated rubric-based judge that has low noise and agrees with human assessment of replication quality. We post-train Faraday, a 27B-parameter "AI Scientist" agent that leverages coding agents as tools, surpassing the performance of Claude Opus 4.8 and GPT-5.5 on held-out replication tasks. Qualitative analysis of individual rollouts reveals that Faraday adopts a more scientifically-principled approach. We believe that our results provide a stepping stone towards AI agents capable of long-horizon scientific innovation without requiring complex harnesses.

Stargazer: A Scalable Model-Fitting Benchmark Environment for AI Agents under Astrophysical Constraints

The rise of autonomous AI agents suggests that dynamic benchmark environments with built-in feedback on scientifically grounded tasks are needed to evaluate the capabilities of these agents in research work. We introduce Stargazer, a scalable environment for evaluating AI agents on dynamic, iterative physics-grounded model-fitting tasks using inference on radial-velocity (RV) time series data. Stargazer comprises 120 tasks across three difficulty tiers, including 20 real archival cases, covering diverse scenarios ranging from high-SNR single-planet systems to complex multi-planetary configurations requiring involved low-SNR analysis. Our evaluation of eight frontier agents reveals a gap between numerical optimization and adherence to physical constraints: although agents often achieve a good statistical fit, they frequently fail to recover correct physical system parameters, a limitation that persists even when agents are equipped with vanilla skills. Furthermore, increasing test-time compute yields only marginal gains, with excessive token usage often reflecting recursive failure loops rather than meaningful exploration. Stargazer presents an opportunity to train, evaluate, scaffold, and scale strategies on a model-fitting problem of practical research relevance today. Our methodology to design a simulation-driven environment for AI agents presumably generalizes to many other model-fitting problems across scientific domains. Source code and the project website are available at https://github.com/Gudmorning2025/Stargazer and https://gudmorning2025.github.io/Stargazer, respectively.

Early science acceleration experiments with GPT-5

AI models like GPT-5 are an increasingly valuable tool for scientists, but many remain unaware of the capabilities of frontier AI. We present a collection of short case studies in which GPT-5 produced new, concrete steps in ongoing research across mathematics, physics, astronomy, computer science, biology, and materials science. In these examples, the authors highlight how AI accelerated their work, and where it fell short; where expert time was saved, and where human input was still key. We document the interactions of the human authors with GPT-5, as guiding examples of fruitful collaboration with AI. Of note, this paper includes four new results in mathematics (carefully verified by the human authors), underscoring how GPT-5 can help human mathematicians settle previously unsolved problems. These contributions are modest in scope but profound in implication, given the rate at which frontier AI is progressing.

Deep Research, information vs. insight, and the nature of science

What AI will accelerate in the scientific process, what it cannot do, and how we can prepare for new manners of scientific investigation.

Denario Project: Deep Knowledge Artificial Intelligence Agents for Scientific Discovery

We present , an Artificial Intelligence (AI) multiagent system designed to serve as a scientific research assistant. can perform many different tasks, such as generating ideas, checking the literature, developing research plans, writing and executing code, making plots, and drafting a scientific paper. The system has a modular architecture, allowing it to handle specific tasks, such as generating an idea or carrying out near end-to-end scientific analysis using as a deep-research backend (end-to-end except for the novelty assessment, which is intentionally left for the user to inspect). In this work, we describe in detail and its modules, and illustrate its capabilities by presenting multiple AI-generated papers generated by it in many different scientific disciplines such as astrophysics, biology, biophysics, biomedical informatics, chemistry, material science, mathematical physics, medicine, neuroscience, and planetary science. can also produce work that combines ideas from different disciplines, which we illustrate with a single ( N = 1 ) example: a paper that applies methods from quantum physics and machine learning to astrophysical data. We report the evaluations performed on these papers by domain experts, who provided both numerical scores and reviewlike feedback. These papers were scored by domain experts on our team; we report these scores as an indicative self-assessment rather than an independent evaluation, and we flag the resulting conflict of interest explicitly. Most drafts were rated above the midpoint of a 0–10 scale, with several rated 8–9, indicating that the generated papers reach the level a good undergraduate or early graduate student could achieve after weeks or a few months of work. They should be regarded as first drafts that require expert development, not publication-ready papers. We also document an important failure mode in which, when the input data cannot be read, the system may silently generate synthetic data, which we flag explicitly as a key limitation. We then highlight the strengths, weaknesses, and limitations of the current system. Finally, we discuss the ethical implications of AI-driven research and reflect on how such technology relates to the philosophy of science. We publicly release the code at . A demo can also be run directly on the web at , and the full app will be deployed on the cloud.

We're releasing Terminal-Bench-Science: a benchmark for evaluating AI agents on research workflows across scientific domains. 🚀 An ongoing community effort out of Stanford University & Laude… | Steven Dillmann | 13 comments

We're releasing Terminal-Bench-Science: a benchmark for evaluating AI agents on research workflows across scientific domains. 🚀 An ongoing community effort out of Stanford University & Laude Institute, built by the team behind Terminal-Bench together with scientific domain experts at research institutions worldwide. 🧪 ⚛️ 🌍 ➕ ⚙️ Terminal-Bench-Science 0.1 has 70 tasks across the life, physical, earth, mathematical, and engineering sciences. Each one is contributed by a researcher porting their own workflows into challenging agentic tasks, on problems they actually care about. 🧠 Anthropic's Opus 5 dominates most scientific domains with an overall pass rate of ~30%, except the mathematical sciences, where Fable 5 and OpenAI's GPT-5.6 Sol take the lead. SpaceXAI's Grok 4.6 is one of the best models for engineering tasks, Z.ai's GLM 5.3 is the best open model, and Kimi (Moonshot AI)'s Kimi K3 is the only model beyond Opus 5 that lands on both cost and token efficiency Pareto frontiers. 🏆 All leaderboard results are live here: https://lnkd.in/gYh99UfU 💻 You can find all tasks on our Github: https://lnkd.in/gejKmXpj Special thanks to our amazing reviewer team Allen Hart, Ivan Bercovich, Joe Janssen, Jiaming H., Steffen Bollmann, Sergey Aganezov, Ben Blaiszik, Karl Krauth, Aaron Feller, Ben Blaiszik, Yiqi Liu, Han Hou, Robert Joseph G., Robert Zhang; our AI research advisors Ryan Marten, Alex Shaw, Lin Shi, Ben Feuer, Mike Merrill, Gian Segato, Jenia Jitsev, Bodhisattwa Majumder, Peter Clark, Thomas Wolf, Braden Hancock, Andy Konwinski, Alex Dimakis; and especially to our co-leads Sanmi Koyejo and Ludwig Schmidt. | 13 comments on LinkedIn

AI co-mathematician: Accelerating mathematicians with agentic AI

We introduce the AI co-mathematician, a workbench for mathematicians to interactively leverage AI agents to pursue open-ended research. The AI co-mathematician is optimized to provide holistic support for the exploratory and iterative reality of mathematical workflows, including ideation, literature search, computational exploration, theorem proving and theory building. By providing an asynchronous, stateful workspace that manages uncertainty, refines user intent, tracks failed hypotheses, and outputs native mathematical artifacts, the system mirrors human collaborative workflows. In early tests, the AI co-mathematician helped researchers solve open problems, identify new research directions, and uncover overlooked literature references. Besides demonstrating a highly interactive paradigm for AI-assisted mathematical discovery, the AI co-mathematician also achieves state of the art results on hard problem-solving benchmarks, including scoring 48% on FrontierMath Tier 4, a new high score among all AI systems evaluated.