Kimi K2.5: Visual Agentic Intelligence

We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This includes a series of techniques such as joint text-vision pre-training, zero-vision SFT, and joint text-vision reinforcement learning. Building on this multimodal foundation, K2.5 introduces Agent Swarm, a self-directed parallel agent orchestration framework that dynamically decomposes complex tasks into heterogeneous sub-problems and executes them concurrently. Extensive evaluations show that Kimi K2.5 achieves state-of-the-art results across various domains including coding, vision, reasoning, and agentic tasks. Agent Swarm also reduces latency by up to $4.5\times$ over single-agent baselines. We release the post-trained Kimi K2.5 model checkpoint to facilitate future research and real-world applications of agentic intelligence.

Balancing Understanding and Generation in Discrete Diffusion Models

In discrete generative modeling, two dominant paradigms demonstrate divergent capabilities: Masked Diffusion Language Models (MDLM) excel at semantic understanding and zero-shot generalization, whereas Uniform-noise Diffusion Language Models (UDLM) achieve strong few-step generation quality, yet neither attains balanced performance across both dimensions. To address this, we propose XDLM, which bridges the two paradigms via a stationary noise kernel. XDLM offers two key contributions: (1) it provides a principled theoretical unification of MDLM and UDLM, recovering each paradigm as a special case; and (2) an alleviated memory bottleneck enabled by an algebraic simplification of the posterior probabilities. Experiments demonstrate that XDLM advances the Pareto frontier between understanding capability and generation quality. Quantitatively, XDLM surpasses UDLM by 5.4 points on zero-shot text benchmarks and outperforms MDLM in few-step image generation (FID 54.1 vs. 80.8). When scaled to tune an 8B-parameter large language model, XDLM achieves 15.0 MBPP in just 32 steps, effectively doubling the baseline performance. Finally, analysis of training dynamics reveals XDLM's superior potential for long-term scaling. Code is available at https://github.com/MzeroMiko/XDLM

EvoClinician: A Self-Evolving Agent for Multi-Turn Medical Diagnosis via Test-Time Evolutionary Learning

Prevailing medical AI operates on an unrealistic ''one-shot'' model, diagnosing from a complete patient file. However, real-world diagnosis is an iterative inquiry where Clinicians sequentially ask questions and order tests to strategically gather information while managing cost and time. To address this, we first propose Med-Inquire, a new benchmark designed to evaluate an agent's ability to perform multi-turn diagnosis. Built upon a dataset of real-world clinical cases, Med-Inquire simulates the diagnostic process by hiding a complete patient file behind specialized Patient and Examination agents. They force the agent to proactively ask questions and order tests to gather information piece by piece. To tackle the challenges posed by Med-Inquire, we then introduce EvoClinician, a self-evolving agent that learns efficient diagnostic strategies at test time. Its core is a ''Diagnose-Grade-Evolve'' loop: an Actor agent attempts a diagnosis; a Process Grader agent performs credit assignment by evaluating each action for both clinical yield and resource efficiency; finally, an Evolver agent uses this feedback to update the Actor's strategy by evolving its prompt and memory. Our experiments show EvoClinician outperforms continual learning baselines and other self-evolving agents like memory agents. The code is available at https://github.com/yf-he/EvoClinician

Dynamic Topology Awareness: Breaking the Granularity Rigidity in Vision-Language Navigation

Vision-Language Navigation in Continuous Environments (VLN-CE) presents a core challenge: grounding high-level linguistic instructions into precise, safe, and long-horizon spatial actions. Explicit topological maps have proven to be a vital solution for providing robust spatial memory in such tasks. However, existing topological planning methods suffer from a "Granularity Rigidity" problem. Specifically, these methods typically rely on fixed geometric thresholds to sample nodes, which fails to adapt to varying environmental complexities. This rigidity leads to a critical mismatch: the model tends to over-sample in simple areas, causing computational redundancy, while under-sampling in high-uncertainty regions, increasing collision risks and compromising precision. To address this, we propose DGNav, a framework for Dynamic Topological Navigation, introducing a context-aware mechanism to modulate map density and connectivity on-the-fly. Our approach comprises two core innovations: (1) A Scene-Aware Adaptive Strategy that dynamically modulates graph construction thresholds based on the dispersion of predicted waypoints, enabling "densification on demand" in challenging environments; (2) A Dynamic Graph Transformer that reconstructs graph connectivity by fusing visual, linguistic, and geometric cues into dynamic edge weights, enabling the agent to filter out topological noise and enhancing instruction adherence. Extensive experiments on the R2R-CE and RxR-CE benchmarks demonstrate DGNav exhibits superior navigation performance and strong generalization capabilities. Furthermore, ablation studies confirm that our framework achieves an optimal trade-off between navigation efficiency and safe exploration. The code is available at https://github.com/shannanshouyin/DGNav.

RL based Beamforming Optimization for 3D Pinching Antenna assisted ISAC Systems

In this paper, a three-dimensional (3D) deployment scheme of pinching antenna array is proposed, aiming to enhances the performance of integrated sensing and communication (ISAC) systems. To fully realize the potential of 3D deployment, a joint antenna positioning, time allocation and transmit power optimization problem is formulated to maximize the sum communication rate with the constraints of target sensing rates and system energy. To solve the sum rate maximization problem, we propose a heterogeneous graph neural network based reinforcement learning (HGRL) algorithm. Simulation results prove that 3D deployment of pinching antenna array outperforms 1D and 2D counterparts in ISAC systems. Moreover, the proposed HGRL algorithm surpasses other baselines in both performance and convergence speed due to the advanced observation construction of the environment.

Resolving Primitive-Sharing Ambiguity in Long-Tailed Industrial Point Cloud Segmentation via Spatial Context Constraints

Industrial point cloud segmentation for Digital Twin construction faces a persistent challenge: safety-critical components such as reducers and valves are systematically misclassified. These failures stem from two compounding factors: such components are rare in training data, yet they share identical local geometry with dominant structures like pipes. This work identifies a dual crisis unique to industrial 3D data extreme class imbalance 215:1 ratio compounded by geometric ambiguity where most tail classes share cylindrical primitives with head classes. Existing frequency-based re-weighting methods address statistical imbalance but cannot resolve geometric ambiguity. We propose spatial context constraints that leverage neighborhood prediction consistency to disambiguate locally similar structures. Our approach extends the Class-Balanced (CB) Loss framework with two architecture-agnostic mechanisms: (1) Boundary-CB, an entropy-based constraint that emphasizes ambiguous boundaries, and (2) Density-CB, a density-based constraint that compensates for scan-dependent variations. Both integrate as plug-and-play modules without network modifications, requiring only loss function replacement. On the Industrial3D dataset (610M points from water treatment facilities), our method achieves 55.74% mIoU with 21.7% relative improvement on tail-class performance (29.59% vs. 24.32% baseline) while preserving head-class accuracy (88.14%). Components with primitive-sharing ambiguity show dramatic gains: reducer improves from 0% to 21.12% IoU; valve improves by 24.3% relative. This resolves geometric ambiguity without the typical head-tail trade-off, enabling reliable identification of safety-critical components for automated knowledge extraction in Digital Twin applications.

T-Mimi: A Transformer-based Mimi Decoder for Real-Time On-Phone TTS

Neural audio codecs provide promising acoustic features for speech synthesis, with representative streaming codecs like Mimi providing high-quality acoustic features for real-time Text-to-Speech (TTS) applications. However, Mimi's decoder, which employs a hybrid transformer and convolution architecture, introduces significant latency bottlenecks on edge devices due to the the compute intensive nature of deconvolution layers which are not friendly for mobile-CPUs, such as the most representative framework XNNPACK. This paper introduces T-Mimi, a novel modification of the Mimi codec decoder that replaces its convolutional components with a purely transformer-based decoder, inspired by the TS3-Codec architecture. This change dramatically reduces on-device TTS latency from 42.1ms to just 4.4ms. Furthermore, we conduct quantization aware training and derive a crucial finding: the final two transformer layers and the concluding linear layers of the decoder, which are close to the waveform, are highly sensitive to quantization and must be preserved at full precision to maintain audio quality.

Improving Methodologies for LLM Evaluations Across Global Languages

As frontier AI models are deployed globally, it is essential that their behaviour remains safe and reliable across diverse linguistic and cultural contexts. To examine how current model safeguards hold up in such settings, participants from the International Network for Advanced AI Measurement, Evaluation and Science, including representatives from Singapore, Japan, Australia, Canada, the EU, France, Kenya, South Korea and the UK conducted a joint multilingual evaluation exercise. Led by Singapore AISI, two open-weight models were tested across ten languages spanning high and low resourced groups: Cantonese English, Farsi, French, Japanese, Korean, Kiswahili, Malay, Mandarin Chinese and Telugu. Over 6,000 newly translated prompts were evaluated across five harm categories (privacy, non-violent crime, violent crime, intellectual property and jailbreak robustness), using both LLM-as-a-judge and human annotation. The exercise shows how safety behaviours can vary across languages. These include differences in safeguard robustness across languages and harm types and variation in evaluator reliability (LLM-as-judge vs. human review). Further, it also generated methodological insights for improving multilingual safety evaluations, such as the need for culturally contextualised translations, stress-tested evaluator prompts and clearer human annotation guidelines. This work represents an initial step toward a shared framework for multilingual safety testing of advanced AI systems and calls for continued collaboration with the wider research community and industry.

Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

AI agents may soon become capable of autonomously completing valuable, long-horizon tasks in diverse domains. Current benchmarks either do not measure real-world tasks, or are not sufficiently difficult to meaningfully measure frontier models. To this end, we present Terminal-Bench 2.0: a carefully curated hard benchmark composed of 89 tasks in computer terminal environments inspired by problems from real workflows. Each task features a unique environment, human-written solution, and comprehensive tests for verification. We show that frontier models and agents score less than 65\% on the benchmark and conduct an error analysis to identify areas for model and agent improvement. We publish the dataset and evaluation harness to assist developers and researchers in future work at https://www.tbench.ai/ .

On the Impact of Phase Errors in Phase-Dependent Amplitudes of Near-Field RISs

This paper investigates mutual coupling between phase-dependent amplitudes (PDAs) and designed phase shifts within pixels of near-field (NF) reconfigurable intelligent surfaces (RISs) in the presence of phase errors (PEs). In contrast to existing research that treats phase shifts with errors (PSEs) and the PDAs separately, we introduce a remaining power (RP) metric to quantify the proportion of power preserved in the signals reflected by the RIS, and we prove its asymptotic convergence to theoretical values by leveraging extended Glivenko-Cantelli theorem. Then, the RP of signals passing through RIS pixels is jointly examined under combined phase and amplitude uncertainties. In addition, we propose four pixel reflection models to capture practical conditions, and we derive approximate polynomial upper bounds for the RP with error terms by applying Taylor expansion. Furthermore, based on Friis transmission formula and projected aperture, we propose a general NF channel model that incorporates the coupling between the PSEs and the PDAs. By using Cauchy-Bunyakovsky-Schwarz inequality and Riemann sums, we derive a closed-form upper bound on spectral efficiency, and the bound becomes tighter as the pixel area decreases. We reveal that as the RIS phase shifts approach the ends of their range, the RP under independent and identically distributed PEs is smaller than that under fully correlated PEs, whereas this relationship reverses when the phase shifts are near the middle of their range. Neglecting the PEs in the PDAs leads to an overestimation of the RIS performance gain, explaining the discrepancies between theoretical and measured results.