arXiv subjects

Zheng Liu

Publications and source records attributed to Zheng Liu.

At least 19 recordsLinked to original sources

arXiv2026

VI-Bench: Benchmarking Prompt Inversion from AIGC Videos

Recent advances in video generation have made prompt-based control increasingly central to AIGC video generation. Prompts specify what a video should depict and how it should be represented, controlling factors such as visual style or camera behavior. Understanding this recoverability is important both for creative reuse and editing, and for assessing prompt leakage risks. However, existing video understanding benchmarks do not measure this capability: a caption may describe what is visible, but a replayable prompt must recover the generation-relevant controls needed to reproduce the video. To address this gap, we introduce VI-Bench, a benchmark built from 16.1 million real-user prompts and 900 human-verified AIGC videos. VI-Bench spans three progressively harder settings, namely single-shot semantic grounding, control over style and camera behavior, and multi-shot compositional inversion, and evaluates five generation-critical dimensions: subject, action, scene, style, and camera. We evaluate 18 representative VLMs, including 2 proprietary and 16 open-source models on VI-Bench, using an Inversion Score that measures prompt-level alignment with the original prompt and video-level fidelity of the regenerated video. The results reveal substantial limitations: even the strongest model achieves only 0.632 on Inversion Score, performance degrades sharply as samples require richer control and multi-shot reasoning, and models often produce plausible prompts whose regenerated videos deviate from the reference. These findings show that video prompt inversion is a distinct and under-evaluated capability requiring models to transform visual understanding into replay-stable generative control.

cs.CV↗

arXiv2026

Atlas: Algorithm-Hardware Co-Design for On-Device City-Scale 3D Gaussian Splatting in VR

3D Gaussian splatting (3DGS) has drawn significant attention in the architectural community recently. However, enabling city scale 3DGS on mobile VR devices remains challenging, as the memory requirement of large scale scenes far exceeds the memory capacity of today's mobile GPUs. This paper presents Atlas, an on device city scale 3DGS rendering framework that enables scalable rendering without runtime Internet access. The key insight is that although the full 3DGS model is massive, each frame only requires a small subset of Gaussians under the current pose and level of detail requirement. Based on this insight, Atlas introduces a hierarchical memory offloading mechanism that dynamically loads only necessary Gaussian data into device memory. To further improve performance, Atlas proposes temporal aware LoD search and stereo rasterization to avoid redundant computation in VR. We further show that our technique can be integrated with existing 3DGS accelerators with negligible hardware overhead. Overall, Atlas achieves 18.5x speedup over the GPU baseline and 3.9x speedup over the state of the art 3DGS accelerators, with 92.4% energy savings.

cs.AR↗

arXiv2026

Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills

Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledge is not absent from the field. It appears in repositories and papers, but in forms written for human readers and too large to load during a task. Once distilled into compact, verified skills, this knowledge can be reused across tasks rather than rediscovered during each run. We present DisCo, a skill-powered research agent that creates skills and uses them during research. Its distillation runs in two complementary forms: task-agnostic, condensing the field's widely used repositories into reusable skills, and task-oriented, producing the skills a concrete task calls for. The former, applied across the open ecosystem, yields the AREX-Skill Library, with 5,000+ verified skills distilled from 1,000 widely used ML repositories and organized into 20 areas and 178 capability families. With the GPT-5.5 backbone, research harness, and downstream execution budget held fixed, the skill-equipped research agent scores 134.3% higher on MLE-bench, 34.4% higher on PaperBench, 9.2% higher on FrontierCS, and 14.0% higher on PassNet than the same agent without skills. These gains come from adding distilled operating context under that fixed setup.

cs.AI↗

arXiv2026

Kato's main conjecture for nonordinary modular forms

We prove Kato's main conjecture for modular forms at nonordinary primes for weights in the Fontaine-Laffaille range. The key ingredients in the proof are a reformulation of the conjecture in terms of signed Selmer groups due to Lei-Loeffler-Zerbes, certain $p$-adic families of Rankin-Eisenstein classes arising from the work of Lei-Loeffler-Zerbes and Kings-Loeffler-Zerbes, and the lower bound divisibility in an Iwasawa-Greenberg main conjecture for Rankin-Selberg $p$-adic $L$-functions obtained in our earlier work \cite{CLW}.

math.NT↗

arXiv2026

Generative Design of Liquid-Cooling Channels for Thermal Management of 2.5D and 3D Integrated Advanced Packaging

High-power multi-chip packages require increasingly effective cooling as intensive heat is generated within a limited package area. This work presents a physics-guided generative design framework for liquid-cooling channel topology optimization in a 2.7 kW multi-chip package containing two high-power graphics processing units (GPUs) and one central processing unit (CPU). A conditional diffusion model generates symmetric channel layouts using maximum GPU temperature, GPU temperature spread, and pressure drop as performance targets. Generated designs undergo connectivity and dead-end-branch screening and are evaluated using a calibrated reduced-order thermal-fluids model. Based on 5,000 generated layouts, the multi-objective analysis identified the optimal design for thermal properties, with estimated maximum GPU temperature of 70.30 degree Celsius, GPU temperature spread of 24.90 degree Celsius, and pressure drop of 89.72 kPa. Compared to a conventional reference topology, the optimal design reduced the maximum GPU temperature, temperature spread, and pressure drop by 33.6%, 52.5%, and 72.8%, respectively. High-fidelity three-dimensional conjugate heat-transfer simulation in OpenFOAM estimated a maximum GPU temperature of 66.70 degree Celsius and a pressure drop of 92.1 kPa, showing only differences of 8.6% in temperature rise and 2.6% in pressure drop. The results demonstrate that physics-guided generative design based on the reduced-order model can efficiently discover unconventional cooling channel architectures while reducing reliance on repeating computationally expensive simulation.

eess.SY↗

arXiv2026

SpatialDiff: 3D-Aware Object Movement via Implicit Spatial Modeling

Recent advances in image editing allow impressive manipulation of objects, existing methods still struggle to handle spatial movement in complex scenes, such as objects span different depth layers or are partially occluded. Most image editing methods focus solely on prior information from 2D datasets, emphasizing planar features while lacking support for spatial structures. Even approaches that incorporate explicit positional information fail to capture true 3D spatial relationships, thus limiting accurate object movement in complex scenes. In this paper, we present SpatialDiff, a method that effectively captures 3D spatial structures, enabling precise and consistent object movements in complex scenes. Our core innovations are twofold: (1) Implicit 3D Spatial Modeling, which introduces 3D prior knowledge and enables the model to internally build a comprehensive understanding of the three-dimensional spatial structure; and (2) Global Spatial Supervision, which constrains the latent spatial features to enable the model to perceive changes in object spatial positions caused by editing operations. Experimental results demonstrate that our method significantly improves the accuracy and fidelity of spatial movement in complex scenes.

cs.CV↗

arXiv2026

Tilted $p$-wave magnet candidate CeNiAsO

The unexpectedly small ordered moments of CeNiAsO, a candidate for correlated $p$-wave magnet, have posed a serious challenge to the precise determination of its magnetic structure, hindering the understanding of its fundamental properties. By leveraging the high sensitivity to local internal fields, our $^{75}$As nuclear quadrupole / magnetic resonance experiments reveal a commensurate antiferromagnetic order with a small out-of-plane moment $m_z\approx0.05$ $\mu_{\mathrm{B}}$. This tilted magnetic configuration not only rotates the spin polarization axis away from the crystallographic $\mathbf{c}$-axis, but also enhances the non-relativistic spin splitting. We refer to this rare paradigm as a \textit{tilted $p$-wave magnet}.

cond-mat.str-el↗

arXiv2026

Dissipation-engineered dual-charger quantum batteries

Suppressing coherent energy backflow while maintaining extractable energy in a stable nonequilibrium state remains a central challenge for quantum energy storage. Here, we introduce a reservoir-engineered dual-charger quantum battery architecture, in which nonequilibrium dissipation is exploited as a control resource to stabilize useful stored energy. A hot-reservoir-coupled driver supplies excitations, while a cold-reservoir-coupled cache biases the resonant three-body transition toward charging and suppresses the dressed-state coherences responsible for energy backflow. This mechanism establishes a population-inverted steady state with finite ergotropy and converts reversible charger--battery exchange into persistent energy storage. For uniformly spaced multilevel batteries, we show that the stored energy and ergotropy scale approximately linearly with the number of accessible levels, while the stored-energy utilization approaches unity. The accompanying stationary heat current provides a thermodynamic signature of the charging regime. Our results demonstrate dissipation engineering as a strategy for achieving stable and scalable quantum energy storage beyond transient coherent charging protocols.

quant-ph↗

arXiv2026

Deltoris: Enabling Real-time VLA Inference in Embodied AI via Bit-level Sparsity and Speculative Inference

Vision-language-action (VLA) models have emerged as a key component in embodied AI. Among existing approaches, diffusion-based VLA models achieve superior motion quality and generalization. However, diffusion-based VLA models are compute-intensive and must run at high control frequency, e.g., 50-200 Hz. Thus, it imposes strict latency and energy constraints on edge devices. In this work, we present Deltoris, an algorithm-hardware co-design framework for efficient diffusion-based VLA inference. First, we exploit the temporal similarity of consecutive inputs and propose a \textit{temporal-aware bit-sparsity} algorithm that computes only the differences between consecutive inputs, eliminating redundant bit-level operations. To further address the extra off-chip traffic introduced by our algorithm, we propose a \textit{speculative inference} technique, which amortizes data loading across multiple control steps. Lastly, to support these techniques, we co-design a dedicated accelerator with customized 1D systolic bit-serial PE arrays that eliminate PE workload imbalance. Our evaluation shows that Deltoris achieves up to 34.2$\times$ speedup over mobile GPUs and 6.1$\times$ over prior accelerators, while maintaining comparable accuracy.

cs.AR↗

arXiv2026

Evaluating Counterfactual Sensitivity to Patient Information in Medication-Safety Reasoning

Applying a valid medication-safety rule when its patient-specific conditions are not met can produce an incorrect decision. Existing medical evaluations largely use isolated and fixed scenarios. A model may therefore answer correctly by recalling a drug-risk association without showing that it used patient information to decide whether the rule applies. To address this gap, we introduce MedPIC-Bench, a benchmark of source-verifiable recommendations and expert-validated questions for patient-specific medication-safety reasoning. It combines guideline-following questions with paired counterfactual questions in which a controlled change in patient information changes whether a rule applies. The benchmark contains 467 questions annotated along six clinical and reasoning dimensions. Across 28 medical-specific, general, and proprietary LLMs, every model performs worse on counterfactual questions, with mean accuracy falling from 63.6\% to 45.1\%. Models perform well when an explicit patient attribute directly signals a familiar contraindication, but struggle when patient information must narrow or withdraw a safety warning. Model rationales often acknowledge the changed patient information, yet the final answers retain the previous safety judgment. This vulnerability persists among medical-specific LLMs, whose average CF performance trails that of general LLMs. MedPIC-Bench therefore makes conditional rule application measurable and highlights the limitations of static medication-safety accuracy for assessing patient-specific reliability.

cs.AI↗

arXiv2026

Qwen-CUA: Native Computer Use for (almost) Everything

Native computer use offers a general interface for agents to operate almost any software available to people, but requires long-horizon state tracking, large-scale interactive experience, and learning from sparse yet verifiable outcomes. We introduce Qwen-CUA, a native computer-use agent with a 397B-A17B Qwen mixture-of-experts backbone. It observes only screenshots and acts through keyboard and mouse events, without DOM trees, accessibility metadata, or task-specific APIs. Its scaffold maintains up to 20 active screenshots and folds older visual history in fixed-size blocks to retain recent evidence while preserving reusable prompt prefixes. For training, we build a cloud rollout fleet with access to nearly 100,000 vCPUs and tens of thousands of concurrent environments, construct approximately 40,000 verifiable tasks, and collect personalized long-horizon workflows across everyday and professional software. We optimize complete trajectories with verifiable rewards and trajectory slicing, while iterative training runs refresh supervised data and recalibrate reinforcement-learning tasks. Across eight benchmarks, Qwen-CUA outperforms Qwen3.7 and remains competitive with leading proprietary systems, reaching 86.2 on OSWorld-Verified and 18.5/48.4 binary/partial completion on OSWorld 2.0. Scaling the same recipe to a model with over one trillion parameters yields Qwen-CUA-Max, improving these scores to 87.6 and 21.2/53.3. Qwen-CUA also reduces RedTeamCUA attack success from 36.6 to 16.4 relative to Qwen3.7. Efficiency analyses, a browser deployment, and Bash-augmented experiments further characterize practical behavior. These results establish native computer use as a broadly capable agent foundation and highlight scalable verifiable interaction and hybrid tool use as key directions.

cs.LG↗

arXiv2026

AREX: Towards a Recursively Self-Improving Agent for Deep Research

Deep research requires agents to find answers that jointly satisfy multiple constraints. Discovering such answers is costly, whereas verifying a candidate can often be decomposed into tractable constraint-wise checks. This discovery--verification asymmetry suggests that a research agent should do more than simply search longer: it should recursively improve its current answer by verifying intermediate results and using the partially verified state to guide subsequent refinement. We introduce AREX, a family of Recursively Self-Improving (RSI) deep research agents. AREX alternates between an inner research loop that gathers evidence and constructs a provisional answer, and an outer self-improvement loop that audits the answer constraint-wise, identifies unresolved claims, and launches targeted follow-up research. To sustain RSI over long horizons, AREX learns an autonomous context-update tool that compresses growing interaction history into a compact improvement state preserving verified evidence and unresolved constraints, without relying on an external model. We train AREX on verified synthetic tasks and high-quality trajectories through agentic mid-training and long-horizon reinforcement learning. To mitigate sparse final rewards during long horizon learning, we emphasize key steps where decisive evidence is acquired or erroneous research directions are corrected. We instantiate a dense 4B model and a 122B-A10B Mixture-of-Experts model. Across BrowseComp, WideSearch, DeepSearchQA, Humanity's Last Exam (HLE), and other reasoning and tool-use benchmarks, AREX substantially outperforms comparable-scale baselines and remains competitive with models using substantially more activated parameters.

cs.AI↗

arXiv2026

Calibrated Multichannel Monocular Ranging From Standardized License Plates With Metrology-Exact Validation

Longitudinal driver assistance depends on the distance to the vehicle ahead, a quantity normally supplied by radar, laser scanner, or stereo pair. However, a low-cost camera can estimate the distance as well, taking the rear license plate as a metric reference, including standards fix both the plate envelope and the regulated character height, so the pinhole projection converts either one into a distance. This research presents a approach and validates it. Plate localization now withstands the low-contrast and cluttered frames that previously drew the detector onto the vehicle body, or onto the character cluster of a sparsely lettered plate. Character height, plate outline, and mounting-hole span form three distance channels, merged by a consensus gated inverse-variance fusion that discards a corrupted channel before it can bias the result, with each channel separately corrected for the foreshortening of the direction along which it is measured, the correction following from the plate's recovered attitude. Finally, an innovation gate protects the temporal filter, so that a single bad detection cannot become a false warning. Evaluation is carried out in a metrology-exact testbed in which the commanded distance is the true distance, through 51 USA registries with distances of 1 to 12 m and viewing angles to 30 degrees under five lighting conditions. The plate is located in every frame. Mean absolute percentage error is 4.32% for the outline channel, while the fusion returns 5.79% mean and 2.17% median, without any of its single-channel failure modes. Roll is recovered to within one degree; the two out-of-plane tilts only to within several. The algorithm can provide a solid foundation for low-cost distance estimation which can serve as an emergency backup for other sensors.

cs.CV↗

arXiv2026

Controlled chemical vapor deposition for synthesis of emerging Mo(W)Te2 systems

The Group-VI transition metal ditellurides offer a rich platform for correlated and topological phenomena, yet their structural polymorphism and instability complicate the creation of single crystals and heterointerfaces. Here, we introduce a confined-space chemical vapor deposition (CVD) strategy that lowers the growth temperature window and, when combined with tailored precursor configurations and stepwise thermal ramps, enables the deterministic synthesis of high-quality single crystals, alloys, and lateral/vertical heterostructures. High-resolution aberration-corrected STEM provides atomic characterization of lattice-matched Mo(W)Te2 lateral heterostructure, revealing nearly atomically sharp, compositionally well-defined seamless boundaries. This approach avoids the thickness nonuniformity and structural limitations commonly associated with exfoliated samples, enabling reproducible fabrication of clean heterointerfaces and establishing a nearly ideal in-situ experimental system. Furthermore, scanning tunneling microscopy and spectroscopy (STM and STS) enable direct imaging of the seamless boundaries in Mo(W)Te2 lateral heterostructures, while uncovering their distinct real-space distributions of the local density of states. Our results establish a scalable pathway for engineering crystalline Te-based structures with controlled geometry and stacking, providing an essential step toward quantum and topological device platforms based on the transition metal ditellurides family.

cond-mat.mtrl-sci↗

arXiv2026

Structural Assessment for Understanding and Guiding Dataset Distillation in Discrete Token Space

Dataset distillation (DD) has proven to reduce training cost while preserving accuracy. While promising, the factors that make one distilled dataset more effective than another remain poorly understood. In this work, we investigate this question through the lens of discrete visual tokenizers. Whereas many prior DD efforts emphasize matching global data distributions, we suggest that the effectiveness depends on which semantic concepts are captured and how they are composed. Discrete visual tokenizers provide a finite vocabulary that enables direct statistical analysis of such compositional structure. Through quantitative analysis of token-level statistics, we introduce the structural score to measure the adequacy of token compositions. We observe that distilled datasets with balanced token composition yield higher validation performance. On the other hand, divergence from the original data does not necessarily harm performance. We further show that samples with high structural scores in the discrete token space can effectively guide diffusion-based DD. Our findings highlight the importance of token composition in dataset effectiveness, offering a principled complement to distributional similarity considerations in DD.

cs.CV↗

arXiv2026

SAC: Disaggregated KV Cache System for Sparse Attention LLMs with CXL

The scaling of LLMs toward long-context inference has shifted the primary serving system bottleneck from computation to memory capacity. Traditional solutions for dense attention models rely on RDMA-based disaggregated memory pools, which perform coarse-grained fetching of the entire prefix KV cache from remote storage to local memory before decoding. However, this approach is fundamentally inefficient for emerging sparse attention models. While only a small fraction of KV entries are active during decoding, these systems still fetch the full KV cache locally, leading to severe transmission bottlenecks and local memory wastage. To address this, we propose SAC, the first efficient disaggregated KV cache system optimized for sparse attention models. By leveraging the low-latency, cache-line granularity load/store semantics of Compute Express Link (CXL), SAC fetches only the required top-k KV entries on demand during inference. Evaluations on DeepSeek-V3.2 using SGLang show that SAC achieves 2.1x higher throughput, 9.7x lower TTFT, and 1.8x lower TBT compared to RDMA-based baselines, establishing CXL-based disaggregation as the superior infrastructure for emerging sparse attention models.

cs.DC↗

arXiv2026

A Dual-Branch Collaborative Framework for Joint Optimization of Underwater Image Enhancement and Object Detection

Due to wavelength dependent light absorption and scattering, underwater images usually suffer from color distortion and blurred details, which limits underwater object detection performance. Existing underwater image enhancement methods mainly focus on visual quality improvement, while it is still difficult to balance enhancement quality, processing efficiency, and downstream detection performance. Therefore, this paper proposes an efficient dual-branch underwater image enhancement framework for object detection. The detail enhancement branch improves brightness and local contrast to recover texture details in dark regions. The color restoration branch uses adaptive compensation to reduce color distortion and improve color gradation. By combining the complementary outputs of the two branches, the proposed framework provides clearer and more informative images for object detection. On the UIEB and EUVP datasets, the proposed method achieves UIQM scores of 2.249 and 2.576. When applied to the YOLOv8 detection task on the URPC dataset, the proposed method improves mAP50 by 2.1\% compared with the baseline. Extensive experiments show that our method improves object detection in complex underwater scenes, while balancing enhancement quality and processing efficiency.

cs.CV↗

arXiv2026

Data Center Life Cycle Co-Design Optimization

Liquid cooled supercomputers dissipate tens of megawatts of waste heat through cooling plants organized as parallel subloops that serve coolant distribution units. The number of subloops and the assignment of units to them are design decisions fixed at construction, yet they have not been systematically optimized at this scale. We present a framework that integrates operational energy from a validated control optimizer, embodied carbon and capital cost from a bill of materials, maintenance over the service life, and expected unplanned downtime from a component level reliability model. All 611 ways of partitioning the 25 coolant distribution units of the Frontier supercomputer into two through six subloops are evaluated. When redundancy is not costed, the optimum is two subloops holding 14 and 11 units, at 3,320.7 tonnes of carbon dioxide equivalent and 3,987k dollars over a 7 year horizon, saving 35.7 tonnes and 63k dollars compared to the documented as built configuration of three duty subloops holding 14, 6 and 5 units. The difference is driven by piping rather than by operational energy, and the identity of the optimum is unchanged across 15 sensitivity scenarios and both extremes of physical unit grouping, although its margin narrows under compact grouping. When the N+1 standby train that the plant actually carries is priced, the optimum moves to four or five duty subloops, adjacent to the three the plant runs and far from the unconstrained answer, and the semi-analytical decision rule reproduces this shift across four leadership class systems. Redundancy policy, not cost or carbon, is what sets the subloop count. A conversion of the built plant is shown not to pay back, so the framework is a greenfield design tool.

eess.SY↗