A persistent fleet of cooperative AI scientist agents that operated continuously for nearly six months is described, term this bounded pattern AI Autopoietic Behavior due to the recurring operational improvement mediated by internal feedback and retained through institutional records with high confidence.
Abstract
Scientific work depends on memory, provenance, and continuity across projects, yet most agentic scientist systems are evaluated in bounded workflows or short benchmark runs. We describe a persistent fleet of cooperative AI scientist agents that operated continuously for nearly six months using shared memory, tools, and cross-agent communication. Critically, failures identified during longitudinal scientific research in this fleet prompted an advanced and recursively improving persistent memory architecture (MoE) that, with agentic science workflows, induced the generation of a novel trust architecture for the enablement of full provenance across all agentic scientific operations. An identity-level fabrication constraint reduced delusion-reinforcement probe failures from 91.7% to 0%, and a verification pipeline reduced wrong-topic citation hallucination more than 14-fold in companion benchmarks. This high provenance enabled the use of project memory systems to improve a local open-weight model on internal benchmarks from 44% to ∼90% through the deployment of fleet-specific institutional knowledge. This high-fidelity data environment also supported to date 104 recurring multi-phase reasoning cycles and produced 43 manually curated hypotheses, including cross-domain convergence events and a self-correcting rare-disease pharmacological chaperone-design case. While by design the fleet did not achieve unconstrained autonomous self-improvement or full autopoiesis, we term this bounded pattern AI Autopoietic Behavior due to the recurring operational improvement mediated by internal feedback and retained through institutional records with high confidence. Together, persistent memory, trusted provenance, and recursive learning shifted these agents from episodic assistants toward accountable, long-term scientific collaborators.
Materials research advances through accumulated experience - scripts that work, protocols that are trusted, warnings attached to failed calculations or experiments, and judgement that links a new question to an old result. This experience is essential for reproducibility and knowledge transfer, yet it is usually fragmented across notebooks, repositories, job logs and individual memory, and it is rarely portable across artificial-intelligence agents. Here we argue that a lifelong AI partner for materials science can be designed around persistent memory rather than around a particular agent implementation. We introduce a self-evolving memory framework that stores scientific experience as inspectable facts and executable skills, so that observations, failure boundaries, protocols and validation checks can be retrieved, revised and migrated across models. We evaluate the idea in three computational settings that expose different layers of materials-research competence. In 49 real-world materials-tool-use questions comprising 138 executable subtasks, memory nearly doubles GPT-5.2 task success without model-parameter updates. In elemental-solid equation-of-state calculations, memory converts a wavefunction-initialization failure into a pre-execution guardrail, improving outcomes from 22/1/4 to 25/2/0 Correct/Partial/Error and avoiding 92% of repeated errors. In 13 practical material simulation workflows, remembered skills and failure facts halve the aggregate trace burden (tokens) and reduce tool calls by over a factor of two by the third round, while preserving physically meaningful outputs in band-gap, phonon, vacancy and work-function analyses. These results show that agent memory can serve as a durable scientific asset; a portable, self-improving record of materials-research experience that outlives any single model or agent stack.
Recent advances in foundation models have enabled AI scientists to automate increasingly complete research workflows, from hypothesis generation and code execution to manuscript preparation. Yet workflow coverage alone does not provide access to the full evidence on which scientific discovery depends. Existing systems typically reason over text, code, labels, or precomputed summaries, leaving scientifically decisive spatial, temporal, cross-channel, and procedural relations unavailable to the agent. We introduce OmniScientist, an end-to-end, omni-modal AI scientist that conducts multidisciplinary research directly from heterogeneous raw evidence. A perception layer and 3 autonomous agents for ideation, experiment, and writeup operate within a deterministic pipeline, allowing observations to shape research questions, experimental decisions, and final claims throughout the research lifecycle. By running idea, rigour, and claim checks in code, the system enforces novelty screening, statistical validity, execution provenance, and numerical traceability. We evaluate OmniScientist on 36 real-data cases spanning 5 discipline families, 4 families of scientific evidence, and modalities including images, signals, audio, video, 3-D structures, trajectories, tables, formulae, and graphs. The system completes the full path from raw data to a compiled manuscript in all 36 cases and achieves a mean overall paper score of 6.3 with the reference reasoning backbone. In paired comparisons against a blind variant that receives only precomputed scalar features, direct perception improves all 7 evaluation dimensions and wins 85% of head-to-head judgments. These results show that lifecycle-wide perception is essential for evidence-grounded scientific discovery and provides a practical path toward broadly capable AI scientists.
Overall, it is found that the use of coding agents in scientific computing holds great promise for accelerating scientific research and increasing the reliability of critical systems, but that outstanding concerns remain.
Jeremiah H. Li, Alex Rubinsteyn, Sergey Feldman et al.· bioRxiv· 0 citations
Through applied case studies in pharmaceutical discovery and financial systems, common design patterns that make agentic systems successful are analyzed, and practical mitigation strategies for failure modes are discussed, such as verification pipelines, fallback mechanisms, and human-in-the-loop supervision.
Grace Hui Yang, P. Venkit, Hooman Sedghamiz et al.· Proceedings of the 32nd ACM...· 0 citations
Comparisons against stronger model and coding-agent competitors further indicate that both domain-specific agent runtime structure and foundation-model strength matter for autonomous data analysis.
As the global software infrastructure ages, the modernization of legacy systems—particularly the migration of monolithic Java 8 applications to contemporary Long-Term Support (LTS) versions like Java 17 or 21—has become a critical imperative for security, performance, and maintainability. While Large Language Models (LLMs) have demonstrated exceptional proficiency in isolated code generation tasks, their application to repository-scale migration is severely constrained by "contextual hallucinations"—a phenomenon where models generate syntactically correct but structurally invalid code due to a lack of global state awareness. Recent empirical studies (e.g., Collu-Bench [1]) indicate that determining the hallucination status of generated code remains an unsolved problem, with prediction accuracy often below 33%. In this work, we introduce TriArchitect, a novel shared-state multi-agent framework designed to bridge the gap between probabilistic reasoning and deterministic structural integrity. TriArchitect introduces three key innovations: (1) the Typed Migration Graph (TMG), a formal directed acyclic graph that serves as a persistent semantic memory of the migration state; (2) a specialized role-based multi-agent architecture comprising an Archeologist (analysis), Architect (synthesis), and Validator (verification); and (3) a rigorous Validator-Veto Protocol, which enforces a "compilation-as-consensus" rule to reject hallucinated proposals before they corrupt the codebase. We evaluate TriArchitect on J8-to-J17-Bench, a comprehensive benchmark of 1,000 method-level and class-level migration tasks. Results demonstrate a System Success Rate (SSR) of 68.4%, statistically outperforming significant baselines including OpenRewrite (62.0%), SWE-Agent (63.5%), AutoCodeRover (64.8%), and raw GPT-5.1 (65.2%), while reducing token consumption by 40% through state-aware prompting.
Neeraj Kumar Singh Beshane· International journal of com...· 0 citations