The Agent Use of Agent Beings: Agent Cybernetics Is the Missing Science of Foundation Agents

LLM-based foundation agents that perceive, reason, and act across thousands of reasoning steps are rapidly becoming the dominant paradigm for deploying artificial intelligence in open-ended, long-horizon complex tasks. Despite this significance, the field remains overwhelmingly engineering-driven. Engineering practice has converged on useful primitives (tool loops, memory banks, harnesses, reflection steps), yet these are assembled by empirical trial and error rather than from first principles. Fundamental questions remain open: under what conditions does a long-running agent remain on-task? How should an agent respond when its environment exceeds its representational capacity? What architectural properties are necessary for safe self-improvement? We argue that cybernetics, the mid-twentieth-century science of control and communication in complex systems, provides the missing theoretical scaffold for foundation agents. By mapping six canonical laws of classical cybernetics onto six agent design principles, and synthesizing those principles into three engineering desiderata (reliability, lifelong running, and self-Improvement), we arrive at a framework termed Agent Cybernetics. Three application domains, code generation, computer use and automated research, exemplify the analytical framework of agent cybernetics by identifying failure modes and concrete engineering recommendations. We hope that agent cybernetics opens a new research venue and establishes the scientific foundation that foundation agents need for principled, reliable real-world deployment.

Interpreting V1 Population Activity via Image-Neural Latent Representation Alignment

Understanding the neural mechanisms underlying visual computation has long been a central challenge in neuroscience. Recent alignment based approaches have improved the accuracy of decoding visual stimuli from brain activity, yet they provide limited insight into the neural computations that give rise to these improvements. To address this gap, we propose Dual-Tower Image-Neural Alignment (DINA), an interpretable contrastive framework for analyzing population level visual computations in primary visual cortex (V1). DINA jointly trains a biologically motivated dual-tower architecture that aligns visual stimuli and corresponding V1 population responses in a shared latent space at the level of intermediate feature maps, enabling both accurate decoding and direct access to interpretable feature maps. Evaluated on large-scale two-photon calcium imaging data from mouse V1, DINA achieves accurate neural-based decoding while revealing that decoding performance is primarily supported by coarse, low-level visual structure, rather than semantic category information or fine-grained details. Further analysis reveals that alignable feature maps emerge from multiple spatially distributed image regions, capturing both shape and texture cues, and are predominantly reconstructed by sparse subsets of strongly responsive neurons and their functional interactions. Together, these results confirm that, beyond enabling accurate decoding, DINA provides a principled framework for probing the computational mechanisms underlying visual processing in V1.

Structure-Augmented Standard Plane Detection with Temporal Aggregation in Blind-Sweep Fetal Ultrasound

In low-resource settings, blind-sweep ultrasound provides a practical and accessible method for identifying fetal growth restriction. However, unlike freehand ultrasound which is subjectively controlled, detection of biometry plane in blind-sweep ultrasound is more challenging due to the uncontrolled fetal structure to be observed and the variaties of oblique planes in the scan. In this work, we propose a structure-augmented system to detect fetal abdomen plane, where the abdominal structure is highlighted using a segmentation prior. Since standard planes are emerging gradually, the decision boundary of the keyframes is unstable to predict. We thus aggregated the structure-augmented planes with a temporal sliding window to help stabilise keyframe localisation. Extensive results indicate that the structure-augmented temporal sliding strategy significantly improves and stabilises the detection of anatomically meaningful planes, which enables more reliable biometric measurements in blind-sweep ultrasound.

HiRO-Nav: Hybrid ReasOning Enables Efficient Embodied Navigation

Embodied navigation agents built upon large reasoning models (LRMs) can handle complex, multimodal environmental input and perform grounded reasoning per step to improve sequential decision-making for long-horizon tasks. However, a critical question remains: \textit{how can the reasoning capabilities of LRMs be harnessed intelligently and efficiently for long-horizon navigation tasks?} In simple scenes, agents are expected to act reflexively, while in complex ones they should engage in deliberate reasoning before acting.To achieve this, we introduce \textbf{H}ybr\textbf{i}d \textbf{R}eas\textbf{O}ning \textbf{Nav}igation (\textbf{HiRO-Nav}) agent, the first kind of agent capable of adaptively determining whether to perform thinking at every step based on its own action entropy. Specifically, by examining how the agent's action entropy evolves over the navigation trajectories, we observed that only a small fraction of actions exhibit high entropy, and these actions often steer the agent toward novel scenes or critical objects. Furthermore, studying the relationship between action entropy and task completion (i.e., Q-value) reveals that improving high-entropy actions contributes more positively to task success.Hence, we propose a tailored training pipeline comprising hybrid supervised fine-tuning as a cold start, followed by online reinforcement learning with the proposed hybrid reasoning strategy to explicitly activate reasoning only for high-entropy actions, significantly reducing computational overhead while improving decision quality. Extensive experiments on the \textsc{CHORES}-$\mathbb{S}$ ObjectNav benchmark showcases that HiRO-Nav achieves a better trade-off between success rates and token efficiency than both dense-thinking and no-thinking baselines.

Graph Topology Information Enhanced Heterogeneous Graph Representation Learning

Real-world heterogeneous graphs are inherently noisy and usually not in the optimal graph structures for downstream tasks, which often adversely affects the performance of GRL models in downstream tasks. Although Graph Structure Learning (GSL) methods have been proposed to learn graph structures and downstream tasks simultaneously, existing methods are predominantly designed for homogeneous graphs, while GSL for heterogeneous graphs remains largely unexplored. Two challenges arise in this context. Firstly, the quality of the input graph structure has a more profound impact on GNN-based heterogeneous GRL models compared to their homogeneous counterparts. Secondly, most existing homogenous GRL models encounter memory consumption issues when applied directly to heterogeneous graphs. In this paper, we propose a novel Graph Topology learning Enhanced Heterogeneous Graph Representation Learning framework (ToGRL).ToGRL learns high-quality graph structures and representations for downstream tasks by incorporating task-relevant latent topology information. Specifically, a novel GSL module is first proposed to extract downstream task-related topology information from a raw graph structure and project it into topology embeddings. These embeddings are utilized to construct a new graph with smooth graph signals. This two-stage approach to GSL separates the optimization of the adjacency matrix from node representation learning to reduce memory consumption. Following this, a representation learning module takes the new graph as input to learn embeddings for downstream tasks. ToGRL also leverages prompt tuning to better utilize the knowledge embedded in learned representations, thus enhancing adaptability to downstream tasks. Extensive experiments on five real-world datasets show that our ToGRL outperforms state-of-the-art methods by a large margin.

MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome

Recent progress in deep research systems has been impressive, but evaluation still lags behind real user needs. Existing benchmarks predominantly assess final reports using fixed rubrics, failing to evaluate the underlying research process. Most also offer limited multimodal coverage, rely on synthetic tasks that do not reflect real-world query complexity, and cannot be refreshed as knowledge evolves. To address these gaps, we introduce MiroEval, a benchmark and evaluation framework for deep research systems. The benchmark comprises 100 tasks (70 text-only, 30 multimodal), all grounded in real user needs and constructed via a dual-path pipeline that supports periodic updates, enabling a live and evolving setting. The proposed evaluation suite assesses deep research systems along three complementary dimensions: adaptive synthesis quality evaluation with task-specific rubrics, agentic factuality verification via active retrieval and reasoning over both web sources and multimodal attachments, and process-centric evaluation audits how the system searches, reasons, and refines throughout its investigation. Evaluation across 13 systems yields three principal findings: the three evaluation dimensions capture complementary aspects of system capability, with each revealing distinct strengths and weaknesses across systems; process quality serves as a reliable predictor of overall outcome while revealing weaknesses invisible to output-level metrics; and multimodal tasks pose substantially greater challenges, with most systems declining by 3 to 10 points. The MiroThinker series achieves the most balanced performance, with MiroThinker-H1 ranking the highest overall in both settings. Human verification and robustness results confirm the reliability of the benchmark and evaluation framework. MiroEval provides a holistic diagnostic tool for the next generation of deep research agents.

Deep Tabular Representation Corrector

Tabular data have been playing a mostly important role in diverse real-world fields, such as healthcare, engineering, finance, etc. The recent success of deep learning has fostered many deep networks (e.g., Transformer, ResNet) based tabular learning methods. Generally, existing deep tabular machine learning methods are along with the two paradigms, i.e., in-learning and pre-learning. In-learning methods need to train networks from scratch or impose extra constraints to regulate the representations which nonetheless train multiple tasks simultaneously and make learning more difficult, while pre-learning methods design several pretext tasks for pre-training and then conduct task-specific fine-tuning, which however need much extra training effort with prior knowledge. In this paper, we introduce a novel deep Tabular Representation Corrector, TRC, to enhance any trained deep tabular model's representations without altering its parameters in a model-agnostic manner. Specifically, targeting the representation shift and representation redundancy that hinder prediction, we propose two tasks, i.e., (i) Tabular Representation Re-estimation, that involves training a shift estimator to calculate the inherent shift of tabular representations to subsequently mitigate it, thereby re-estimating the representations and (ii) Tabular Space Mapping, that transforms the above re-estimated representations into a light-embedding vector space via a coordinate estimator while preserves crucial predictive information to minimize redundancy. The two tasks jointly enhance the representations of deep tabular models without touching on the original models thus enjoying high efficiency. Finally, we conduct extensive experiments on state-of-the-art deep tabular machine learning models coupled with TRC on various tabular benchmarks which have shown consistent superiority.

Unifying Stable Optimization and Reference Regularization in RLHF

Reinforcement Learning from Human Feedback (RLHF) has advanced alignment capabilities significantly but remains hindered by two core challenges: \textbf{reward hacking} and \textbf{stable optimization}. Current solutions independently address these issues through separate regularization strategies, specifically a KL-divergence penalty against a supervised fine-tuned model ($π_0$) to mitigate reward hacking, and policy ratio clipping towards the current policy ($π_t$) to promote stable alignment. However, the implicit trade-off arising from simultaneously regularizing towards both $π_0$ and $π_t$ remains under-explored. In this paper, we introduce a unified regularization approach that explicitly balances the objectives of preventing reward hacking and maintaining stable policy updates. Our simple yet principled alignment objective yields a weighted supervised fine-tuning loss with a superior trade-off, which demonstrably improves both alignment results and implementation complexity. Extensive experiments across diverse benchmarks validate that our method consistently outperforms RLHF and online preference learning methods, achieving enhanced alignment performance and stability.

Multi-Scale Wavelet Transformers for Operator Learning of Dynamical Systems

Recent years have seen a surge in data-driven surrogates for dynamical systems that can be orders of magnitude faster than numerical solvers. However, many machine learning-based models such as neural operators exhibit spectral bias, attenuating high-frequency components that often encode small-scale structure. This limitation is particularly damaging in applications such as weather forecasting, where misrepresented high frequencies can induce long-horizon instability. To address this issue, we propose multi-scale wavelet transformers (MSWTs), which learn system dynamics in a tokenized wavelet domain. The wavelet transform explicitly separates low- and high-frequency content across scales. MSWTs leverage a wavelet-preserving downsampling scheme that retains high-frequency features and employ wavelet-based attention to capture dependencies across scales and frequency bands. Experiments on chaotic dynamical systems show substantial error reductions and improved long horizon spectral fidelity. On the ERA5 climate reanalysis, MSWTs further reduce climatological bias, demonstrating their effectiveness in a real-world forecasting setting.

Causal Preference Elicitation

We propose causal preference elicitation, a Bayesian framework for expert-in-the-loop causal discovery that actively queries local edge relations to concentrate a posterior over directed acyclic graphs (DAGs). From any black-box observational posterior, we model noisy expert judgments with a three-way likelihood over edge existence and direction. Posterior inference uses a flexible particle approximation, and queries are selected by an efficient expected information gain criterion on the expert's categorical response. Experiments on synthetic graphs, protein signaling data, and a human gene perturbation benchmark show faster posterior concentration and improved recovery of directed effects under tight query budgets.