arXiv:2606. 12848v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for tasks once reserved for trained researchers, including hypothesis generation, specification choice, and drafting conclusions.
arXiv:2607. 21268v1 Announce Type: cross Abstract: In many social-science research tasks, such as economics, LLM-based agents must produce outputs for which no cheap, task-complete, machine-readable correctness signal exists.
The paper introduces Atria Dawn Preview, a foundation agentic language model aimed at scientific research and engineering workflows. Trained through a Verifiable Experience Pipeline, it performs competitively across 16 real‑world benchmarks, achieving the highest scores on five. The authors also present a detailed case study of human–AI collaboration, showing that while AI proposes methods and revisions, humans retain final decision‑making and guide the research direction.
The paper investigates the limitations of post-training AI agents that can autonomously train large language models. It distinguishes between execution-level capability—making adjustments within a chosen training strategy—and strategy-level capability—revising the overall approach based on new evidence. Analysis of many public post-training runs shows that agents lock into a strategy early and then only perform local tweaks, regardless of task. Experiments with experience scaffolds, human guidance, and extra compute improve execution but do not enable strategy reevaluation, indicating that agents lack a mechanism to spontaneously reassess their strategy during training.
arXiv:2609.07611v1 Announce Type: new Abstract: Scientific ideation is the capacity to formulate novel and testable hypotheses from scientific evidence, and autonomous AI scientists depend on it. Exi...
arXiv:2606. 07462v1 Announce Type: new Abstract: As foundation models advance and agent scaffolding becomes increasingly sophisticated, agents have demonstrated remarkable proficiency in complex, long-horizon coding tasks and even autonomous experiment execution.
arXiv:2606. 11217v1 Announce Type: cross Abstract: The proliferation of large language models (LLMs) and autonomous AI agents has given rise to a rapidly growing methodological paradigm: "in silico" behavioral experiments.
The paper argues that large language models need adaptive reasoning rather than fixed reasoning budgets. It shows that over‑reasoning leads to high computational cost without accuracy gains, while under‑reasoning results in incorrect or incomplete solutions. The authors evaluate these failure modes on MATH‑500 and the GAIA benchmark, highlighting the need for dynamic reasoning allocation in agentic AI systems.
ASI‑Bench is a new benchmark that evaluates AI systems on their ability to conduct innovative exploration and autonomous scientific research across 11 domains, using 60 project‑level tasks. It progressively removes human methodological guidance to test whether AI can independently select methods, execute research, and produce verifiable results. Results from 18 state‑of‑the‑art agent–model configurations show a sharp performance drop when guidance is reduced, indicating current systems still rely heavily on human input.
We study what happens when a single general-purpose large language model acts as the sole researcher on a long-horizon neural architecture design problem. The agent receives a scientific question, an initial hypothesis and motivation, a compute budget, and research affordances (source and experiment management, experiment tracking, literature access, and persistent memory), then autonomously proposes, implements, evaluates, and records experiments over an extended period.
arXiv:2606. 00376v1 Announce Type: new Abstract: Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks, not due to preference biases, but limits rooted in the information-theoretic capacity of decoder-only attention.
arXiv:2607. 27191v1 Announce Type: cross Abstract: Forecasts of explosive AI progress hinge on AI agents automating AI research.
The paper introduces CoSLR, a Human‑AI collaborative system for systematic literature reviews that incorporates mandatory human checkpoints within a three‑phase pipeline using large language models and Retrieval‑Augmented Generation. In a survey of 63 participants, 42.9 % rated the system’s usability highly, yet 34.9 % indicated they would trust AI‑generated summaries without further human verification after brief interaction. The study highlights that effective human oversight in AI‑assisted literature reviews depends on users’ willingness to engage with the checkpoints, underscoring a calibration issue that interface design must directly address.