arXiv subjects

Yao Xu

Publications and source records attributed to Yao Xu.

At least 19 recordsLinked to original sources

arXiv2026

Interactive World Simulation with Cosmos for Physical AI

We present a shared framework that adapts Cosmos world foundation models for real-time interactive world simulation in Physical AI. By streaming visual observations in response to actions and scene controls, the framework enables users and embodied agents to interact with evolving virtual environments. It supports both predicting the visual consequences of actions and generating observations from physical state maintained by an external simulator. We demonstrate this approach across camera navigation and robot manipulation, controllable video transfer, and multi-view autonomous vehicle (AV) simulation. These examples provide a shared foundation for developing interactive virtual environments for Physical AI training and evaluation. Our code and model checkpoints are available at https://github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3-examples. The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos-simulation.

cs.AI↗

arXiv2026

AskStream: When and What to Ask Proactively under Uncertainty in Streaming Video Understanding

Recent advances in multimodal large language models have increased interest in interaction paradigms for video understanding. The evolution from offline to streaming and proactive interaction has gradually shifted models from passive responders toward more proactive agents that can decide when to answer. However, existing proactive paradigms typically assume that the information required to answer a query is complete, leaving models to only wait or answer. In the real world, however, information is often incomplete due to uncertainty from either the video or the user. In proactive paradigms, such uncertainty can cause the model to wait forever or generate highly hallucinatory responses. We therefore argue that proactive interaction should be redefined with a third action: ask, allowing models to actively ask for missing information. To study when and what models should ask before answering, we introduce AskStream, the first multi-turn benchmark spanning video and user uncertainty for streaming video understanding, with four task categories and eight subtasks. It contains 438 videos with 2,714 manually annotated frame-level time windows across three types: wait, ask, and answer. Across both multiple-choice and open-ended settings, even the strongest model attains only 52.18% on our proposed Ask-F1 metric, highlighting substantial limitations in proactive information seeking under uncertainty.

cs.CV↗

arXiv2026

Coordinating Boundaries: Improved Approximation Algorithms for Unweighted Non-Preemptive Throughput Maximization

We study unweighted nonpreemptive throughput maximization: given jobs with processing times, release times, and deadlines, schedule a maximum-cardinality subset on identical machines. For every fixed $\varepsilon>0$, we give randomized algorithms with expected approximation ratios $17/13+\varepsilon$ in polynomial time and $21/17+\varepsilon$ in pseudopolynomial time. These improve the respective $4/3+\varepsilon$ and $5/4+\varepsilon$ guarantees of Armbruster, Grandoni, Tinguely, and Wiese [STOC'26]. Both results hold for an arbitrary number of identical machines. Our algorithms build on their configuration-LP and matching framework, which partitions time into blocks. For a job whose window meets multiple blocks, its boundary blocks are the first and last such blocks. Crossing boundary pairs, whose endpoints strictly interleave, permit additional assignments to sampled execution slots, while a noncrossing family can be coordinated using consistent joint distributions on at most three blocks. The exchange analysis accounts for fractional job assignments that must be erased before inserting an additional job. For the pseudopolynomial implementation, we project configurations onto selected job identities and use separate configuration distributions with common placement marginals. This retains efficient pricing while enforcing the correlations needed for coordination.

cs.DS↗

arXiv2026

RuleTrek: Benchmarking Agentic Rule Reasoning in Complex Scenarios

Large language models (LLMs) increasingly tackle complex real-world tasks in specialized domains. However, identifying and correctly applying domain-specific rules remains challenging when scenario information is distributed across multiple files. Existing rule-reasoning benchmarks typically represent scenarios using only a few factual statements, simplifying the identification and application of relevant rules. To address this gap, this paper introduces \textbf{RuleTrek}, a multi-domain benchmark for agentic rule reasoning in complex scenarios. Its rule-grounded construction pipeline extracts structured rule graphs, generates and incrementally composes rule-specific sub-scenarios while ensuring consistency and solvability. To systematically evaluate agentic rule reasoning in complex scenarios, RuleTrek assesses three capabilities in interactive workspaces: Rule Identification (RI), Rule Deduction (RD), and End-to-End Rule Reasoning (E2E). Experiments on 11 frontier LLMs show that the highest RI and RD scores are only 62.47\% and 60.43\%, respectively, while E2E scores range from 26.97\% to 55.29\%. These results demonstrate that current LLM agents face substantial challenges in rule reasoning over complex scenarios. We make our code and dataset publicly available at \href{https://github.com/yuan247/RuleTrek}{RuleTrek}.

cs.AI↗

arXiv2026

Breaking the $\sqrt{3}$ Barrier for Maximum Weighted $3$-Set Packing

We give a deterministic polynomial-time $1.6908$-approximation for Maximum Weighted $3$-Set Packing, breaking the $\sqrt3$ locality-gap barrier of squared-weight local search. The approximation ratio for this problem progressed from Berman's $2$ [Ber00] to Neuwohner's $2-\frac{1}{63{,}700{,}992}+ε$ [Neu21]. Thiery and Ward then obtained $1.786$ [TW23], while Thiery subsequently improved the bound to $1.761+ε$ and finally to $\sqrt3 \approx 1.732051$ through a layered exchange analysis [Thi23]. Thiery also proved that $\sqrt3$ is a locality-gap lower bound for the squared-weight objective even with exchanges of arbitrary size. Our algorithm performs in two phases and combines two objectives. Phase~I computes a bounded-exchange local optimum for the squared-weight potential and analyzes it through Thiery's layered framework, while strengthening the terminal analysis by preserving internal tree-edge slack for nonsingleton components and exploiting the incidence structure of $3$-sets for final singletons. This yields an augmented structural inequality with residual positive claw gain under the original objective. Phase~II switches to the original objective and recovers sufficient residual gain through an auxiliary weighted $9$-Set Packing instance. A covering argument transfers the structural bound through the high-girth lift used only in the analysis.

cs.DS↗

arXiv2026

Cross-Shift Analysis for Unrelated-Machine Weighted Completion Time A (1.3168+epsilon)-Approximation

We study the problem of minimizing total weighted completion time on unrelated parallel machines with machine-independent job weights. The best previous approximation guarantee for this problem is arbitrarily close to 1.36, due to Li [SODA 2025], who developed a configuration-LP and iterative-rounding framework based on randomly shifted geometric size classes and a computer-assisted final analysis. We improve the approximation guarantee to arbitrarily close to 1.3168. The scheduling algorithm retains Li's configuration-LP and iterative-rounding framework, with a different fixed geometric class ratio. The improvement comes from a new analysis of the random geometric shift. For each machine and Smith prefix, we normalize the physical prefix before averaging over the shift, so that its normalized size distribution and configuration moments remain fixed across all shifts. A scale-dependent configuration bound and a cross-shift averaging argument then reduce the approximation analysis to a one-dimensional certificate. The final certificate is computer assisted and reproducible. Its rational data are verified exactly where possible, while the remaining continuous inequalities are certified using directed interval arithmetic. This yields the first improvement over Li's guarantee for the machine-independent-weight model.

cs.DS↗

arXiv2026

GraphWalker: Agentic Knowledge Graph Question Answering via Synthetic Trajectory Curriculum

Agentic knowledge graph question answering (KGQA) requires an agent to iteratively interact with knowledge graphs (KGs), posing challenges in both training data scarcity and reasoning generalization. Specifically, existing approaches often restrict agent exploration: prompting-based methods lack autonomous navigation training, while current training pipelines usually confine reasoning to predefined trajectories. To this end, this paper proposes GraphWalker, a novel agentic KGQA framework that addresses these challenges through Automated Trajectory Synthesis and Stage-wise Fine-tuning. GraphWalker adopts a two-stage SFT training paradigm: First, the agent is trained on structurally diverse trajectories synthesized from constrained random-walk paths, establishing a broad exploration prior over the KG; Second, the agent is further fine-tuned on a small set of expert trajectories to develop reflection and error recovery capabilities. Extensive experiments demonstrate that our stage-wise SFT paradigm unlocks a higher performance ceiling for a lightweight reinforcement learning (RL) stage, enabling GraphWalker to achieve state-of-the-art performance on CWQ and WebQSP. Additional results on GrailQA and our constructed GraphWalkerBench confirm that GraphWalker enhances generalization to out-of-distribution reasoning paths. The code is publicly available at https://github.com/XuShuwenn/GraphWalker.

cs.CL↗

arXiv2026

MMOU: A Massive Multi-Task Omni Understanding and Reasoning Benchmark for Long and Complex Real-World Videos

Multimodal Large Language Models (MLLMs) have shown strong performance in visual and audio understanding when evaluated in isolation. However, their ability to jointly reason over omni-modal (visual, audio, and textual) signals in long and complex videos remains largely unexplored. We introduce MMOU, a new benchmark designed to systematically evaluate multimodal understanding and reasoning under these challenging, real-world conditions. MMOU consists of 20,000 carefully curated questions paired with 11877 web-collected videos of varying length, spanning diverse domains and exhibiting rich, tightly coupled audio-visual content. The benchmark covers 13 fundamental skill categories, all of which require integrating evidence across modalities and time. All questions are manually annotated across multiple turns by professional annotators, ensuring high quality and reasoning fidelity. We evaluate 20+ state-of-the-art open-source and proprietary multimodal models on MMOU. The results expose substantial performance gaps: the best closed-source model achieves only 64.2% accuracy, while the strongest open-source model reaches just 46.8%. Our results highlight the challenges of long-form omni-modal understanding, revealing that current models frequently fail to apply even fundamental skills in long videos. Through detailed analysis, we further identify systematic failure modes and provide insights into where and why current models break.

cs.CL↗

arXiv2026

Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence

We introduce Nemotron 3 Nano Omni, the latest model in the Nemotron multimodal series and the first to natively support audio inputs alongside text, images, and video. Nemotron 3 Nano Omni delivers consistent accuracy improvements over its predecessor, Nemotron Nano V2 VL, across all modalities, enabled by advances in architecture, training data and recipes. In particular, Nemotron 3 delivers leading results in real-world document understanding, long audio-video comprehension, and agentic computer use. Built on the highly efficient Nemotron 3 Nano 30B-A3B backbone, Nemotron 3 Nano Omni further incorporates innovative multimodal token-reduction techniques to deliver substantially lower inference latency and higher throughput than other models of similar size. We are releasing model checkpoints in BF16, FP8, and FP4 formats, along with portions of the training data and codebase to facilitate further research and development.

cs.LG↗

arXiv2026

Quantitative homogenization of elliptic equations with infinitely many scales

In this paper, we develop a general homogenization theory for elliptic equations with coefficients that oscillate periodically at infinitely many scales $\varepsilon = (\varepsilon_1, \varepsilon_2, \cdots) \in (0,1)^\infty$, with $\varepsilon_1>\varepsilon_2>\cdots$ and $\varepsilon_n \to 0$ as $n \to \infty$. Such problems arise naturally in the study of fractal materials and diffusion in fluids. Under suitable scale-separation assumptions, we prove a qualitative homogenization theorem and obtain optimal $L^2$ convergence rates. We also establish interior and boundary Lipschitz estimates that are uniform in $\varepsilon$.

math.AP↗

arXiv2026

TiMem: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents

Long-horizon conversational agents have to manage ever-growing interaction histories that quickly exceed the finite context windows of large language models (LLMs). Existing memory frameworks provide limited support for temporally structured information across hierarchical levels, often leading to fragmented memories and unstable long-horizon personalization. We present TiMem, a temporal--hierarchical memory framework that organizes conversations through a Temporal Memory Tree (TMT), enabling systematic memory consolidation from raw conversational observations to progressively abstracted persona representations. TiMem is characterized by three core properties: (1) temporal--hierarchical organization through TMT; (2) semantic-guided consolidation that enables memory integration across hierarchical levels without fine-tuning; and (3) complexity-aware memory recall that balances precision and efficiency across queries of varying complexity. Under a consistent evaluation setup, TiMem achieves state-of-the-art accuracy on both benchmarks, reaching 75.30% on LoCoMo and 76.88% on LongMemEval-S. It outperforms all evaluated baselines while reducing the recalled memory length by 52.20% on LoCoMo. Manifold analysis indicates clear persona separation on LoCoMo and reduced dispersion on LongMemEval-S. Overall, TiMem treats temporal continuity as a first-class organizing principle for long-horizon memory in conversational agents. The code is available at https://github.com/TiMEM-AI/timem.

cs.CL↗

arXiv2026

Uni-Animator: Towards Unified Visual Colorization

We propose Uni-Animator, a novel Diffusion Transformer (DiT)-based framework for unified image and video sketch colorization. Existing sketch colorization methods struggle to unify image and video tasks, suffering from imprecise color transfer with single or multiple references, inadequate preservation of high-frequency physical details, and compromised temporal coherence with motion artifacts in large-motion scenes. To tackle imprecise color transfer, we introduce visual reference enhancement via instance patch embedding, enabling precise alignment and fusion of reference color information. To resolve insufficient physical detail preservation, we design physical detail reinforcement using physical features that effectively capture and retain high-frequency textures. To mitigate motion-induced temporal inconsistency, we propose sketch-based dynamic RoPE encoding that adaptively models motion-aware spatial-temporal dependencies. Extensive experimental results demonstrate that Uni-Animator achieves competitive performance on both image and video sketch colorization, matching that of task-specific methods while unlocking unified cross-domain capabilities with high detail fidelity and robust temporal consistency.

cs.CV↗

arXiv2026

A comparison of different image analysis techniques for mapping spatiotemporal pH and carbon dissolution in density-driven convection of CO2 in water

Density-driven convection enhances the carbon dissolution rate, which is significant for the geological carbon storage. This process will also influence the spatiotemporal pH and carbon concentrations of the underground fluid. To illuminate the convection mechanism, it is critical to understand the evolution of those properties within the porous media. However, determining the spatiotemporal pH and concentration within porous media is always challenging. This study employed a combination of three pH indicators that can track a wide range in pH from 4 to 9.5 in a convection experiment. Furthermore, we compared three image-processing techniques: Hue, gray-difference, and angular representation of RGB color space $\mathbf{(ϕ,θ)}$ for quantifying color changes from the universal $\text{pH}$ indicator arising from the carbon convection. The characterized colors were mapped into pH by calibrating against benchmark solutions. The comparative results demonstrate that the color quantified by the Hue technique is most robust, showing invariance to fluid thickness, camera settings, and LED luminance. In the convection experiments, it produces a continuous spatial distribution of pH and concentration level in the system. In contrast, the $\mathbf{(ϕ,θ)}$ and gray-difference techniques were more sensitive to environmental variations. They also have significant limitations for $\text{pH}$ interpolation in the critical range due to their non-monotonic calibration paths. Although all methods ultimately produced similar estimates of total dissolved carbon, the Hue technique offers greater stability and universality for high-resolution, dynamic measurements of pH and carbon concentration in the convection experiments.

physics.flu-dyn↗

arXiv2026

Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems

Enhancing customer experience is essential for business success, particularly as service demands grow in scale and complexity. Generative artificial intelligence and Large Language Models (LLMs) have empowered intelligent interaction systems to deliver efficient, personalized, and 24/7 support. In practice, intelligent interaction systems encounter several challenges: (1) Constructing high-quality data for cold-start training is difficult, hindering self-evolution and raising labor costs. (2) Multi-turn dialogue performance remains suboptimal due to inadequate intent understanding, rule compliance, and solution extraction. (3) Frequent evolution of business rules affects system operability and transferability, constraining low-cost expansion and adaptability. (4) Reliance on a single LLM is insufficient in complex scenarios, where the absence of multi-agent frameworks and effective collaboration undermines process completeness and service quality. (5) The open-domain nature of multi-turn dialogues, lacking unified golden answers, hampers quantitative evaluation and continuous optimization. To address these challenges, we introduce WOWService, an intelligent interaction system tailored for industrial applications. With the integration of LLMs and multi-agent architectures, WOWService enables autonomous task management and collaborative problem-solving. Specifically, WOWService focuses on core modules including data construction, general capability enhancement, business scenario adaptation, multi-agent coordination, and automated evaluation. Currently, WOWService is deployed on the Meituan App, achieving significant gains in key metrics, e.g., User Satisfaction Metric 1 (USM 1) -27.53% and User Satisfaction Metric 2 (USM 2) +25.51%, demonstrating its effectiveness in capturing user needs and advancing personalized service.

cs.CL↗

arXiv2025

NVIDIA Nemotron Parse 1.1

We introduce Nemotron-Parse-1.1, a lightweight document parsing and OCR model that advances the capabilities of its predecessor, Nemoretriever-Parse-1.0. Nemotron-Parse-1.1 delivers improved capabilities across general OCR, markdown formatting, structured table parsing, and text extraction from pictures, charts, and diagrams. It also supports a longer output sequence length for visually dense documents. As with its predecessor, it extracts bounding boxes of text segments, as well as corresponding semantic classes. Nemotron-Parse-1.1 follows an encoder-decoder architecture with 885M parameters, including a compact 256M-parameter language decoder. It achieves competitive accuracy on public benchmarks making it a strong lightweight OCR solution. We release the model weights publicly on Huggingface, as well as an optimized NIM container, along with a subset of the training data as part of the broader Nemotron-VLM-v2 dataset. Additionally, we release Nemotron-Parse-1.1-TC which operates on a reduced vision token length, offering a 20% speed improvement with minimal quality degradation.

cs.LG↗

arXiv2025

NVIDIA Nemotron Nano V2 VL

We introduce Nemotron Nano V2 VL, the latest model of the Nemotron vision-language series designed for strong real-world document understanding, long video comprehension, and reasoning tasks. Nemotron Nano V2 VL delivers significant improvements over our previous model, Llama-3.1-Nemotron-Nano-VL-8B, across all vision and text domains through major enhancements in model architecture, datasets, and training recipes. Nemotron Nano V2 VL builds on Nemotron Nano V2, a hybrid Mamba-Transformer LLM, and innovative token reduction techniques to achieve higher inference throughput in long document and video scenarios. We are releasing model checkpoints in BF16, FP8, and FP4 formats and sharing large parts of our datasets, recipes and training code.

cs.LG↗

arXiv2025

Towards Agentic Self-Learning LLMs in Search Environment

We study whether self-learning can scale LLM-based agents without relying on human-curated datasets or predefined rule-based rewards. Through controlled experiments in a search-agent setting, we identify two key determinants of scalable agent training: the source of reward signals and the scale of agent task data. We find that rewards from a Generative Reward Model (GRM) outperform rigid rule-based signals for open-domain learning, and that co-evolving the GRM with the policy further boosts performance. Increasing the volume of agent task data-even when synthetically generated-substantially enhances agentic capabilities. Building on these insights, we propose \textbf{Agentic Self-Learning} (ASL), a fully closed-loop, multi-role reinforcement learning framework that unifies task generation, policy execution, and evaluation within a shared tool environment and LLM backbone. ASL coordinates a Prompt Generator, a Policy Model, and a Generative Reward Model to form a virtuous cycle of harder task setting, sharper verification, and stronger solving. Empirically, ASL delivers steady, round-over-round gains, surpasses strong RLVR baselines (e.g., Search-R1) that plateau or degrade, and continues improving under zero-labeled-data conditions, indicating superior sample efficiency and robustness. We further show that GRM verification capacity is the main bottleneck: if frozen, it induces reward hacking and stalls progress; continual GRM training on the evolving data distribution mitigates this, and a small late-stage injection of real verification data raises the performance ceiling. This work establishes reward source and data scale as critical levers for open-domain agent learning and demonstrates the efficacy of multi-role co-evolution for scalable, self-improving agents. The data and code of this paper are released at https://github.com/forangel2014/Towards-Agentic-Self-Learning

cs.AI↗

arXiv2025

Cognitive-Aligned Spatio-Temporal Large Language Models For Next Point-of-Interest Prediction

The next point-of-interest (POI) recommendation task aims to predict the users' immediate next destinations based on their preferences and historical check-ins, holding significant value in location-based services. Recently, large language models (LLMs) have shown great potential in recommender systems, which treat the next POI prediction in a generative manner. However, these LLMs, pretrained primarily on vast corpora of unstructured text, lack the native understanding of structured geographical entities and sequential mobility patterns required for next POI prediction tasks. Moreover, in industrial-scale POI prediction applications, incorporating world knowledge and alignment of human cognition, such as seasons, weather conditions, holidays, and users' profiles (such as habits, occupation, and preferences), can enhance the user experience while improving recommendation performance. To address these issues, we propose CoAST (Cognitive-Aligned Spatial-Temporal LLMs), a framework employing natural language as an interface, allowing for the incorporation of world knowledge, spatio-temporal trajectory patterns, profiles, and situational information. Specifically, CoAST mainly comprises of 2 stages: (1) Recommendation Knowledge Acquisition through continued pretraining on the enriched spatial-temporal trajectory data of the desensitized users; (2) Cognitive Alignment to align cognitive judgments with human preferences using enriched training data through Supervised Fine-Tuning (SFT) and a subsequent Reinforcement Learning (RL) phase. Extensive offline experiments on various real-world datasets and online experiments deployed in "Guess Where You Go" of AMAP App homepage demonstrate the effectiveness of CoAST.

cs.AI↗