arXiv:2605.24636v2 Announce Type: new Abstract: While large language models (LLMs) hold transformative potential for medicine, their reasoning robustness and safety in real-world clinical scenarios remain critically underexplored, particularly in dentistry. Here we introduce GlobalDentBench, the first multinational dental benchmark, featuring a taxonomy that encompasses 14 dental specialties across 88 countries and regions spanning six continents. The benchmark comprises 8,978 expert-validated questions across three formats (multiple-choice, short-answer, and case-based questions) and assesses three progressive reasoning levels: knowledge recall (L1), routine reasoning (L2), and individualized reasoning (L3). To ensure data quality, the automated construction framework was calibrated by six senior dentists, achieving expert agreement rates of 99.98% for multiple-choice and short-answer questions and 96.78% for the more complex case-based questions. Evaluation of 12 frontier LLMs on GlobalDentBench revealed a sharp, stepwise performance degradation with increasing reasoning complexity. Specifically, accuracy plummeted from 81.34% on multiple-choice to 64.53% on short-answer and 22.34% on case-based questions, while declining markedly from 74.01% at L1 to 55.64% at L2 and 35.71% at L3. More critically, risk analysis of real-world dental cases demonstrated an alarming overall unsafe rate of 31.01% in LLM-generated clinical recommendations, with 4.51% posing risks of irreversible patient harm and risks particularly pronounced in specialties such as orthodontics. These findings expose fundamental limitations in the medical reasoning and safety of current LLMs. Consequently, GlobalDentBench provides a scalable foundation for trustworthy clinical AI evaluation, underscoring the urgent need for rigorous validation before the safe deployment of these models in healthcare.

arXiv:2511.15407v4 Announce Type: replace Abstract: Humans learn by observing, interacting with environments, and internalizing physics and causality. Here, we aim to ask whether an agent can similarly acquire human-like reasoning from interaction and keep improving with more experience. To study this, we introduce a Game-to-Unseen (G2U) benchmark of 1,000+ heterogeneous games that exhibit significant visual domain gaps. Existing approaches, including VLMs and world models, struggle to capture underlying physics and causality since they are not focused on core mechanisms and overfit to visual details. VLM/VLA agents reason but lack look-ahead in interactive settings, while world models imagine but imitate visual patterns rather than analyze physics and causality. We therefore propose IPR (Interactive Physical Reasoner), using world-model rollouts to score and reinforce a VLM's policy, and introduce PhysCode, a physics-centric action code aligning semantic intent with dynamics to provide a shared action space for prediction and reasoning. Pretrained on 1,000+ games, our IPR performs robustly on levels from primitive intuition to goal-driven reasoning, and even surpasses GPT-5 overall. We find that performance improves with more training games and interaction steps, and that the model also zero-shot transfers to unseen games. These results support physics-centric interaction as a path to steadily improving physical reasoning. Further demos and project details can be found at https://mybearyzhang.github.io/ipr-1.

arXiv:2602.22769v3 Announce Type: replace Abstract: Large Language Models (LLMs) are deployed as autonomous agents in increasingly complex applications, where enabling long-horizon memory is critical for achieving strong performance. However, a significant gap exists between applications and evaluation standards for agent memory: existing benchmarks primarily focus on dialogue-centric settings. In reality, agent memory consists of a continuous stream of agent-environment interactions that are primarily composed of machine-generated representations. To bridge this gap, we introduce AMA-Bench (Agent Memory with Any Length), a benchmark designed to evaluate long-horizon memory for LLMs in real agentic applications. It features two key components: (1) a set of real-world agentic trajectories across representative agentic applications, paired with expert-curated QA, and (2) a set of synthetic agentic trajectories of arbitrary horizons paired with rule-based QA. Our comprehensive study shows that existing memory systems underperform on AMA-Bench primarily because they lack causality and objective information, and are constrained by the lossy nature of similarity-based retrieval employed by many memory systems. To address these limitations, we propose AMA-Agent, an effective memory system featuring a causality graph and tool-augmented retrieval. Our results demonstrate that AMA-Agent achieves 57.22% average accuracy on AMA-Bench, surpassing the strongest baselines by 11.16%. Resources are available at our project website: https://ama-bench.github.io/

arXiv:2505.05375v3 Announce Type: replace-cross Abstract: Recently, spiking neural networks (SNNs), deployed on neuromorphic chips, provide highly efficient solutions on edge devices in different scenarios. However, their ability to adapt to distribution shifts after deployment has become a crucial challenge. Online test-time adaptation (OTTA) offers a promising solution by enabling models to dynamically adjust to new data distributions without requiring source data or labeled target samples. Nevertheless, existing OTTA methods are largely designed for traditional artificial neural networks and are not well-suited for SNNs. To address this gap, we propose a low-power, neuromorphic chip-friendly online test-time adaptation framework, aiming to enhance model generalization under distribution shifts. The proposed approach is called Threshold Modulation (TM), which dynamically adjusts the firing threshold through neuronal dynamics-inspired normalization, being more compatible with neuromorphic hardware. Experimental results on benchmark datasets demonstrate the effectiveness of this method in improving the robustness of SNNs against distribution shifts while maintaining low computational cost. The proposed method offers a practical solution for online test-time adaptation of SNNs, providing inspiration for the design of future neuromorphic chips. The demo code is available at github.com/NneurotransmitterR/TM-OTTA-SNN.

arXiv:2603.24239v2 Announce Type: replace-cross Abstract: Dynamism is common in AI computation, e.g., the dynamic tensor shapes and the dynamic control flows in models. Due to the long compilation time, existing runtime compilation damages the model efficiency, while the offline compilers either suffer from the long compilation time and device memory footprint to cover all the possible execution instances of a dynamic model, or sacrifice optimization opportunities for usability. In this paper, we rethink the feasibility of runtime compilation for dynamic models and identify that the key for it to work is to speed up the compilation or hide the compilation overhead. To do this, we propose a real-time compiler, DVM. In DVM, we design a runtime operator compiler based on a bytecode virtual machine to perform effective and efficient compilation for each dynamic operator instance given its input. Specifically, instead of compiling programs into machine code, we encode the operator program into bytecode on the CPU and decode the bytecode into virtual instructions for direct execution on the NPU. Based on the runtime operator compiler, we further propose an operator fuser, which performs symbol-deduction-based fusion on static graphs and runtime fusion on dynamic graphs. Both pattern- and stacking-based fusion are supported to increase fusion opportunities. Evaluation on operators, subgraphs, and models shows that, compared with TorchInductor, PyTorch-eager and MindSpore-graph-O0, we are up to 11.77$\times$ better in terms of the operator/model efficiency and up to 5 orders of magnitude faster in terms of the maximum compilation time.

arXiv:2603.10062v2 Announce Type: replace-cross Abstract: As LLM agents evolve into collaborative multi-agent systems, their memory requirements grow rapidly in complexity. This position paper frames multi-agent memory as a computer architecture problem. We distinguish shared and distributed memory paradigms, propose a three-layer memory hierarchy (I/O, cache, and memory), and identify two critical protocol gaps: cache sharing across agents and structured memory access control. We argue that the most pressing open challenge is multi-agent memory consistency. Our architectural framing provides a foundation for building reliable, scalable multi-agent systems.

arXiv:2603.06757v1 Announce Type: cross Abstract: Unlabeled streaming data are usually collected to describe dynamic systems, where concept drift detection is a vital prerequisite to understanding the evolution of systems. However, the drifting concepts are usually imbalanced in most real cases, which brings great challenges to drift detection. That is, the dominant statistics of large clusters can easily mask the drifting of small cluster distributions (also called small concepts), which is known as the `masking effect'. Considering that most existing approaches only detect the overall existence of drift under the assumption of balanced concepts, two critical problems arise: 1) where the small concept is, and 2) how to detect its drift. To address the challenging concept drift detection for imbalanced data, we propose Imbalanced Cluster Descriptor-based Drift Detection (ICD3) approach that is unbiased to the imbalanced concepts. This approach first detects imbalanced concepts by employing a newly designed multi-distribution-granular search, which ensures that the distribution of both small and large concepts is effectively captured. Subsequently, it trains a One-Cluster Classifier (OCC) for each identified concept to carefully monitor their potential drifts in the upcoming data chunks. Since the detection is independently performed for each concept, the dominance of large clusters is thus circumvented. ICD3 demonstrates highly interpretability by specifically locating the drifted concepts, and is robust to the changing of the imbalance ratio of concepts. Comprehensive experiments with multi-aspect ablation studies conducted on various benchmark datasets demonstrate the superiority of ICD3 against the state-of-the-art counterparts.

arXiv:2603.03297v1 Announce Type: cross Abstract: Test-time Training enables model adaptation using only test questions and offers a promising paradigm for improving the reasoning ability of large language models (LLMs). However, it faces two major challenges: test questions are often highly difficult, making self-generated pseudo-labels unreliable, and existing methods lack effective mechanisms to adapt to a model's specific reasoning weaknesses, leading to inefficient learning. To address these issues, we propose \textbf{TTSR}, a self-reflective test-time self-evolving training framework. TTSR employs a single pretrained language model that alternates between the roles of a \textit{Student} and a \textit{Teacher} at test time. The Student focuses on solving problems and learning from synthesized variant questions, while the Teacher analyzes the Student's failed reasoning trajectories, summarizes recurring reasoning weaknesses, and synthesizes targeted variant questions accordingly. This process guides the model to improve within a learnable regime through a continual self-evolving loop. Experimental results on multiple challenging mathematical reasoning benchmarks show that TTSR consistently improves reasoning performance and generalizes well across different model backbones and general-domain reasoning tasks. These findings suggest that teacher-mediated self-reflection provides an effective pathway for stable and continual reasoning improvement at test time.

arXiv:2603.03387v1 Announce Type: cross Abstract: Clustering is a fundamental approach to understanding data patterns, wherein the intuitive Euclidean distance space is commonly adopted. However, this is not the case for implicit cluster distributions reflected by qualitative attribute values, e.g., the nominal values of attributes like symptoms, marital status, etc. This paper, therefore, discovered a tree-like distance structure to flexibly represent the local order relationship among intra-attribute qualitative values. That is, treating a value as the vertex of the tree allows to capture rich order relationships among the vertex value and the others. To obtain the trees in a clustering-friendly form, a joint learning mechanism is proposed to iteratively obtain more appropriate tree structures and clusters. It turns out that the latent distance space of the whole dataset can be well-represented by a forest consisting of the learned trees. Extensive experiments demonstrate that the joint learning adapts the forest to the clustering task to yield accurate results. Comparisons of 10 counterparts on 12 real benchmark datasets with significance tests verify the superiority of the proposed method.

arXiv:2602.22769v2 Announce Type: replace Abstract: Large Language Models (LLMs) are deployed as autonomous agents in increasingly complex applications, where enabling long-horizon memory is critical for achieving strong performance. However, a significant gap exists between practical applications and current evaluation standards for agent memory: existing benchmarks primarily focus on dialogue-centric, human-agent interactions. In reality, agent memory consists of a continuous stream of agent-environment interactions that are primarily composed of machine-generated representations. To bridge this gap, we introduce AMA-Bench (Agent Memory with Any length), which evaluates long-horizon memory for LLMs in real agentic applications. It features two key components: (1) a set of real-world agentic trajectories across representative agentic applications, paired with expert-curated QA, and (2) a set of synthetic agentic trajectories that scale to arbitrary horizons, paired with rule-based QA. Our comprehensive study shows that existing memory systems underperform on AMA-Bench primarily because they lack causality and objective information and are constrained by the lossy nature of similarity-based retrieval employed by many memory systems. To address these limitations, we propose AMA-Agent, an effective memory system featuring a causality graph and tool-augmented retrieval. Our results demonstrate that AMA-Agent achieves 57.22% average accuracy on AMA-Bench, surpassing the strongest memory system baselines by 11.16%.