Discovery Foundation Models: Toward Open-Ended Discovery Intelligence

• URL: http://arxiv.org/abs/2609.15973v1
• Date: Mon, 14 Sep 2026 17:56:00 GMT
• Title: Discovery Foundation Models: Toward Open-Ended Discovery Intelligence
• Abstract summary: We formulate Discovery Foundation Models (DFMs) as general-purpose model systems for open-ended discovery.<n>A DFM operates over a revisable research state and supports seven coupled capabilities spanning problem discovery, formulation, representation construction, hypothesis formation, intervention, evidence-grounded revision, and continual discovery improvement.<n>We ground the framework with GALILEO, a real therapeutic-discovery system in which Dry-Lab reasoning, robotic and hands-on Wet-Lab experimentation, external biological evidence, and iterative hypothesis and design revision form a closed physical discovery loop.
• Score: 12.795168111957343
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Foundation models have progressed from learning and reasoning over existing knowledge, to increasingly learning through action, tool use, and outcome feedback. We argue that the next frontier is a further transition: from solving and acting within problems specified by humans to participating in the process by which new problems, representations, explanations, and knowledge are created. We refer to this capability as Discovery Intelligence. We formulate Discovery Foundation Models (DFMs) as general-purpose model systems for open-ended discovery. A DFM operates over a revisable research state and supports seven coupled capabilities spanning problem discovery, formulation, representation construction, hypothesis formation, intervention, evidence-grounded revision, and continual discovery improvement. We instantiate this framework with Zetema, which couples explicit research-state dynamics, verification and experimental gating, external grounding, and cross-task Discovery Skill evolution. We further ground the framework with GALILEO, a real therapeutic-discovery system in which Dry-Lab reasoning, robotic and hands-on Wet-Lab experimentation, external biological evidence, and iterative hypothesis and design revision form a closed physical discovery loop. We then formulate a unified approach to capability formation and process-centered evaluation, enabling discovery behavior to be trained, improved, and measured beyond final-answer performance. Together, these components establish discovery as a learnable, executable, and evaluable capability of foundation-model systems. We view this shift as a broader progression in intelligence scaling: from learning over existing knowledge, to learning from action outcomes, and ultimately to participating in the construction, testing, and revision of the structures through which new knowledge is discovered. Code: https://github.com/Gen-Verse/DFM-Plans

Related papers

• From Observation to Insight: Mechanistic World Models and the Quest for Autonomous Discovery [60.11725132433485]
  We argue that scientific discovery is fundamentally a problem of knowledge organisation.<n>We introduce Mechanistic World Models, a new design paradigm that places reusable mechanisms at the centre of representation, computation and learning.
  arXiv  Detail & Related papers  (2026-07-14T08:03:27Z)
• Intentmaking and Sensemaking: Human Interaction with AI-Guided Mathematical Discovery [72.49593499512092]
  We identify and characterize a distinct workflow we term intentmaking.<n>We frame this as a natural extension to sensemaking.<n>Our documentation of these themes suggests an approach to designing AI tools for scientific discovery.
  arXiv  Detail & Related papers  (2026-05-07T09:30:25Z)
• Theory of Space: Can Foundation Models Construct Spatial Beliefs through Active Exploration? [83.13508919229939]
  Theory of Space is defined as an agent's ability to actively acquire information through self-directed, active exploration.<n>A key innovation is spatial belief probing, which prompts models to reveal their internal spatial representations at each step.<n>Our findings suggest that current foundation models struggle to maintain coherent, revisable spatial beliefs during active exploration.
  arXiv  Detail & Related papers  (2026-02-04T19:06:40Z)
• Accelerating Scientific Research with Gemini: Case Studies and Common Techniques [105.15622072347811]
  Large language models (LLMs) have opened new avenues for accelerating scientific research.<n>We present a collection of case studies demonstrating how researchers have successfully collaborated with advanced AI models.
  arXiv  Detail & Related papers  (2026-02-03T18:56:17Z)
• Unlearning as Ablation: Toward a Falsifiable Benchmark for Generative Scientific Discovery [0.0]
  Do large language models (LLMs) truly generate new knowledge, or do they merely remix memorized fragments?<n>We propose unlearning-as-ablation as a falsifiable probe of constructive scientific discovery.
  arXiv  Detail & Related papers  (2025-08-25T05:24:15Z)
• From Reasoning to Learning: A Survey on Hypothesis Discovery and Rule Learning with Large Language Models [3.861418246476623]
  In pursuit of artificial general intelligence (AGI), there is a growing need for models that learn, reason, and generate new knowledge.<n>This survey offers a structured lens to examine Large Language Models-based hypothesis discovery.<n>We synthesize existing work in hypothesis generation, application, and validation, identifying both key achievements and critical gaps.
  arXiv  Detail & Related papers  (2025-05-28T03:40:02Z)
• SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoning [0.0]
  A key challenge in artificial intelligence is the creation of systems capable of autonomously advancing scientific understanding. We present SciAgents, an approach that leverages three core concepts. The framework autonomously generates and refines research hypotheses, elucidating underlying mechanisms, design principles, and unexpected material properties. Our case studies demonstrate scalable capabilities to combine generative AI, ontological representations, and multi-agent modeling, harnessing a swarm of intelligence' similar to biological systems.
  arXiv  Detail & Related papers  (2024-09-09T12:25:10Z)
• Open-world machine learning: A review and new outlooks [117.33922838201993]
  Article presents a holistic view of open-world machine learning.<n>It investigates unknown rejection, novelty discovery, and continual learning.<n>It aims to help researchers build more powerful AI systems in their respective fields.
  arXiv  Detail & Related papers  (2024-03-04T06:25:26Z)
• Anti-Retroactive Interference for Lifelong Learning [65.50683752919089]
  We design a paradigm for lifelong learning based on meta-learning and associative mechanism of the brain. It tackles the problem from two aspects: extracting knowledge and memorizing knowledge. It is theoretically analyzed that the proposed learning paradigm can make the models of different tasks converge to the same optimum.
  arXiv  Detail & Related papers  (2022-08-27T09:27:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.