Searcharxiv⌕ Search

arXiv subjects

Jun Li

Publications and source records attributed to Jun Li.

At least 91 records · Page 5Linked to original sources
arXiv2026

Self-Creative Text-to-Object Generation using Semantic-Aware Spatial Weighting

Instilling creativity in text-to-image (T2I) generation presents a significant challenge, as it requires synthesized images to exhibit not only visual novelty and surprise, but also artistic value. Current T2I models, however, are largely optimized for literal text-image alignment with their data distribution, and their noise prediction networks constrain the generation to high-probability regions, consequently generating outputs that lack authentic creativity. To address this, we propose a Self-Creative Diffusion (SCDiff) model for meaningful T2I generations featuring two core modules: a learnable spatial weighting (LSW) module and a visual-semantic mixing loss (VSML). The LSW module designs a parametric Kaiser-Bessel window to reinforce central image features, fostering novel and surprising generation. The VSML module introduces a dual loss function: a similarity loss constrains that the new images align with its textual description, while a diversity loss maximizes its distinction from the original image, enhancing both semantic value and visual novelty. Extensive experiments demonstrate that our model substantially improves creativity, semantic alignment, and visual coherence, offering a simple yet powerful framework for generating creative objects.

cs.CV↗
arXiv2026

Bézier Degradation Modeling for LiDAR-based Human Motion Capture

LiDAR-based 3D human motion capture has broad applications in fields such as autonomous driving and robotics, where accurate motion reconstruction is crucial. However, existing methods often struggle with unstable inputs and severe occlusions, leading to jittery or even failed pose predictions. To address these challenges, we propose BMLiCap, a coarse-to-fine framework that models motion using temporally compressible Bézier curves. By reducing control points through a trajectory-preserving strategy, we obtain a coherent and learning-friendly motion representation. To reconstruct human actions from LiDAR point-cloud cues, we design a progressive motion-reconstruction module. Specifically, a Time-scale Motion Transformer (TMT) is introduced to predict motion curves at multiple temporal scales, and a Multi-level Motion Aggregator (MMA) is utilized to adaptively fuse the multi-scale curves to recover detailed, temporally coherent poses, effectively bridging observation gaps caused by occlusions and noise. Across four mainstream benchmarks LiDARHuman26M, FreeMotion, NoiseMotion, and SLOPER4D, BMLiCap achieves state-of-the-art accuracy and temporal continuity in complex scenes, demonstrating its ability to compensate for severe occlusions and reduce prediction jitter.

cs.CV↗
arXiv2026

CAR-SAM: Cross-Attention Reconstruction for Post-Training Quantization of the Segment Anything Model

Segment Anything Models (SAMs) are extensively used in computer vision for universal image segmentation, but deploying them on resource-constrained devices is challenging due to their high computational and memory demands. Post-Training Quantization (PTQ) is a widely used technique for model compression and acceleration. However, existing PTQ methods fail to consider the cross-attention architecture in the SAM decoder. This degradation primarily stems from the unique challenges posed by SAMs: (1) Attention dissipation, where the attention information in the decoder, which is crucial for representing segmentation masks, collapses into a diffuse and non-semantic form under low-bit quantization; and (2) Reconstruction oscillation, where bidirectional coupling within the two-way transformer introduces cross-branch error interference and destabilizes convergence. To tackle these issues, we propose CAR-SAM, a unified quantization framework tailored for SAMs. Firstly, to mitigate attention dissipation, we introduce MatMul-Aware Compensation (MAC) mechanism that transfers activation-induced quantization errors from MatMul to preceding linear weights. Secondly, to mitigate oscillation in decoder optimization, we develop a Joint Cross-Attention Reconstruction (JCAR) strategy that jointly reconstructs coupled attention branches, suppressing oscillatory behavior and promoting stable convergence. Extensive experiments show that CAR-SAM robustly quantizes SAM models down to 4-bit precision, surpassing existing methods by 14.6% and 6.6% mAP on SAM-B and SAM-L respectively.

cs.CV↗
arXiv2026

A Heterogeneous Dual-Network Framework for Emergency Delivery UAVs: Communication Assurance and Path Planning Coordination

Natural disasters often damage ground infrastructure, making unmanned aerial vehicles (UAVs) essential for emergency supply delivery. Yet safe operation in complex post-disaster environments requires reliable command-and-control (C2) links; link instability can cause loss of control, delay rescue, and trigger severe secondary harm. To provide continuous three-dimensional (3D) C2 coverage during dynamic missions, we propose a Heterogeneous Dual-Network Framework (HDNF) for safe and reliable emergency delivery. HDNF tightly couples an Emergency Communication Support Network (ECSN), formed by hovering UAV base stations, with a Delivery Path Network (DPN), formed by fast-moving delivery UAVs. The ECSN dynamically safeguards mission-critical flight corridors, while the DPN aligns trajectories with reliable coverage regions. We formulate a joint optimization problem over task assignment, 3D UAV-BS deployment, and DPN path planning to maximize end-to-end C2 reliability while minimizing UAV flight energy consumption and base-station deployment cost. To solve this computationally intractable NP-hard problem, we develop a layered strategy with three components: (i) a multi-layer C2 service model that overcomes 2D-metric limitations and aligns UAV-BS deployment with mission-critical 3D phases; (ii) a 3D coverage-aware multi-agent reinforcement learning algorithm that addresses the high-dimensional search space and improves both training efficiency and topology resilience; and (iii) a 3D communication-aware A* planner that jointly optimizes C2 quality and flight energy, mitigating trajectory--coverage mismatch and improving routing safety. Extensive simulations show that HDNF markedly improves C2 reliability, eliminates outages in critical phases, and sustains high task success rates while reducing hardware deployment cost.

cs.IT↗
arXiv2026

RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation

Intensive care units (ICU) generate long, dense and evolving streams of clinical information, where physicians must repeatedly reassess patient states under time pressure, underscoring a clear need for reliable AI decision support. Existing ICU benchmarks typically treat historical clinician actions as ground truth. However, these actions are made under incomplete information and limited temporal context of the underlying patient state, and may therefore be suboptimal, making it difficult to assess the true reasoning capabilities of AI systems. We introduce RealICU, a hindsight-annotated benchmark for evaluating large language models (LLMs) under realistic ICU conditions, where labels are created after senior physicians review the full patient trajectory. We formulate four physician-motivated tasks: assess Patient Status, Acute Problems, Recommended Actions, and Red Flag actions that risk unsafe outcomes. We partition each trajectory with 30-min windows and release two datasets: RealICU-Gold with 930-window annotations from 94 MIMIC-IV patients, and RealICU-Scale with 11,862 windows extended by Oracle, a physician-validated LLM hindsight labeler. Existing LLMs including memory-augmented ones performed poorly on RealICU, exposing two failure modes: a recall-safety tradeoff for clinical recommendations, and an anchoring bias to early interpretations of the patient. We further introduce ICU-Evo to study structured-memory agents that improves long-horizon reasoning but does not fully eliminate safety failures. Together, RealICU provides a clinically grounded testbed for measuring and improving AI sequential decision-support in high-stakes care. Project page: https://chengzhi-leo.github.io/RealICU-Bench/

cs.AI↗
arXiv2026

Multi-Scale Dequant: Eliminating Dequantization Bottleneck via Activation Decomposition for Efficient LLM Inference

Quantization is essential for efficient large language model (LLM) inference, yet the dequantization step-converting low-bit weights back to high-precision for matrix multiplication has become a critical bottleneck on modern AI accelerators. On architectures with decoupled compute units (e.g., Ascend NPUs), dequantization operations can consume more cycles than the matrix multiplication itself, leaving the high-throughput tensor cores underutilized. This paper presents Multi-Scale Dequant (MSD), a quantization framework that removes weight/KV dequantization from the GEMM critical path. Instead of lifting low-bit weights to BF16 precision, MSD decomposes high-precision BF16 activations into multiple low-precision components, each of which can be multiplied directly with quantized weights via native hardware-accelerated GEMM. This approach shifts the computational paradigm from precision conversion to multi-scale approximation, avoiding INT8-to-BF16 weight conversion before GEMM. We instantiate MSD for two weight formats and derive tight error bounds for each. For INT8 weights (W4A16), two-pass INT8 decomposition achieves near 16 effective bits. For MXFP4 weights (W4A16), two-pass MXFP4 decomposition yields near 6.6 effective bits with error bound 1/64 per block surpassing single-pass MXFP8(5.24 bits) while maintaining the same effective GEMM compute time. We further derive closed-form latency and HBM traffic models showing that MSD avoids the Vector-Cube pipeline stall caused by dequantization and reduces KV cache HBM traffic by up to 2.5 times in attention. Numerical simulations on matrix multiplication and Flash Attention kernels confirm that MSD does not degrade accuracy compared to dequantization baselines, and in many settings achieves lower L2 error.

stat.ML↗
arXiv2026

U-STS-LLM A Unified Spatio-Temporal Steered Large Language Model for Traffic Prediction and Imputation

The efficient operation of modern cellular networks hinges on the accurate analysis of spatio-temporal traffic data. Mastering these patterns is essential for core network functions, chiefly forecasting future load to pre-empt congestion and imputing missing values caused by sensor failures or transmission errors to ensure data continuity. While deeply connected, forecasting and imputation have historically evolved as separate sub-fields. The dominant paradigm, Spatio-Temporal Graph Neural Networks (STGNNs), while effective, are often specialized, computationally intensive, and exhibit limited generalization. Concurrently, adapting large pre-trained language models (LLMs) offers a powerful alternative for sequence modeling, yet existing approaches provide weak structural guidance, leading to unstable convergence and a narrow focus on forecasting. To bridge these gaps, we propose U-STS-LLM, a unified framework built on a spatio-temporally steered LLM. Our core innovation is a Dynamic Spatio-Temporal Attention Bias Generator that synthesizes a persistent functional graph with transient nodal states to explicitly steer the LLM's attention. Coupled with a partially frozen backbone tuned via Low-Rank Adaptation (LoRA) and a Gated Adaptive Fusion mechanism, the model achieves stable, parameter-efficient adaptation. Trained under a unified multi-task objective, U-STS-LLM learns a holistic data representation. Extensive experiments on real-world cellular datasets demonstrate that U-STS-LLM establishes new state-of-the-art performance in both long-horizon forecasting and high-missing-rate imputation, while maintaining remarkable training efficiency and stability, offering a novel blueprint for harnessing foundation models in structured, non-linguistic domains.

cs.LG↗
arXiv2026

The double splitting iteration method for solving the large indefinite least squares problem

Addressing large-scale indefinite least squares (ILS) problem poses notable computational bottlenecks in the field of numerical linear algebra. State-of-the-art iterative schemes for such problems are predominantly constructed upon the single splitting of the coefficient matrix derived from the corresponding normal equation. In this work, we put forward an innovative iterative framework grounded in the double splitting of normal equations tailored for ILS problem. Specifically, we elaborate on a distinct implementations of the double splitting strategy, which offer constructive insights and methodological references for subsequent research on double splitting-based iterative methods. Two numerical experiments further corroborate that the proposed double splitting iterative paradigm outperforms conventional single splitting approaches in both computational efficiency and convergence robustness.

math.NA↗
arXiv2026

Revisiting Uncertainty: On Evidential Learning for Partially Relevant Video Retrieval

Partially relevant video retrieval aims to retrieve untrimmed videos using text queries that describe only partial content. However, the inherent asymmetry between brief queries and rich video content inevitably introduces uncertainty into the retrieval process. In this setting, vague queries often induce semantic ambiguity across videos, a challenge that is further exacerbated by the sparse temporal supervision within videos, which fails to provide sufficient matching evidence. To address this, we propose Holmes, a hierarchical evidential learning framework that aggregates multi-granular cross-modal evidence to quantify and model uncertainty explicitly. At the inter-video level, similarity scores are interpreted as evidential support and modeled via a Dirichlet distribution. Based on the proposed three-fold principle, we perform fine-grained query identification, which then guides query-adaptive calibrated learning. At the intra-video level, to accumulate denser evidence, we formulate a soft query-clip alignment via flexible optimal transport with an adaptive dustbin, which alleviates sparse temporal supervision while suppressing spurious local responses. Extensive experiments demonstrate that Holmes outperforms state-of-the-art methods. Code is released at https://github.com/lijun2005/ICML26-Holmes.

cs.CV↗
arXiv2026

Probing the scalar-induced gravitational waves with the Five-hundred-meter Aperture Spherical radio Telescope and the Square Kilometer Array

Gravitational wave astronomy presents a promising opportunity to directly observe scalar-induced gravitational waves originating from the early universe. Experiments, including ground-based interferometers like LIGO and Virgo, and the Pulsar Timing Array, such as FAST and SKA, are poised to significantly enhance sensitivity to these gravitational waves. In this paper, we combined Cosmic Microwave Background and Baryon Acoustic Oscillation datasets with upper or lower limits of the stochastic gravitational wave background provided by FAST or SKA, to constrain scalar-induced gravitational waves. To provide a comprehensive forecast, we consider two scenarios at a frequency: one where FAST or SKA does not detect scalar-induced gravitational waves, thereby setting an upper limit on the fractional energy density; and another where these waves are detected successfully, thus establishing a lower limit. In the $Λ$CDM+$r$ model, the scalar spectral index of the power-law power spectrum is constrained to $n_s=0.9598^{+0.0013}_{-0.0009}$ from the combinations of CMB+BAO+SKA datasets in the upper limit scenario where scalar-induced gravitational waves propagate at the speed of light. The constraint shifts to $n_s = 0.9697\pm{0.0033}$ in the lower limit scenario. Comparing with the constraint from the combinations of CMB+BAO datasets, the scalar spectral index $n_s$ in the upper limit scenario exhibits significant changes, which could serve as an indicator for detecting scalar-induced gravitational waves. In the $Λ$CDM+$α_s$+$r$ model and $Λ$CDM+$α_s$+$β_s$+$r$ model, the running of the scalar spectral index $α_s$ and the running of the running $β_s$ also show notable variations, suggesting potential indicators. The numerical findings clearly demonstrate the impact of the upper and lower limits provided by FAST or SKA.

astro-ph.CO↗
arXiv2026

Dynamic Decision Learning: Test-Time Evolution for Abnormality Grounding in Rare Diseases

Clinical abnormality grounding for rare diseases is often hindered by data scarcity, making supervised fine-tuning impractical and single-pass inference highly unstable. We propose Dynamic Decision Learning (DDL), a framework that enables frozen large vision-language models (LVLMs) to refine their decisions across both language and visual spaces by optimizing instructions and consolidating predictions under visual perturbations. This process improves localization quality and produces a consensus-based reliability score that quantifies model confidence. Results on brain imaging benchmarks, including a rare-disease dataset with 281 pathology types across models ranging from 3B to 72B parameters, show that DDL improves mAP@75 by up to 105% on rare-disease cases and outperforms adaptation baselines and supervised fine-tuning. Furthermore, DDL demonstrates stronger calibration between reliability scores and localization accuracy under severe distribution shifts and increasing task difficulty. Code is available at: https://lijunrio.github.io/DDL/

cs.CL↗
arXiv2026

Thinking Like a Clinician: A Cognitive AI Agent for Clinical Diagnosis via Panoramic Profiling and Adversarial Debate

The application of large language models (LLMs) in clinical decision support faces significant challenges of "tunnel vision" and diagnostic hallucinations present in their processing unstructured electronic health records (EHRs). To address these challenges, we propose a novel chain-based clinical reasoning framework, called DxChain, which transforms the diagnostic workflow into an iterative process by mirroring a clinician's cognitive trajectory that consists of "Memory Anchoring", "Navigation" and "Verification" phases. DxChain introduces three key methodological innovations to elicit the potential of LLM: (i) a Profile-Then-Plan paradigm to mitigate cold-start hallucinations by establishing a panoramic patient baseline, (ii) a Medical Tree-of-Thoughts (Med-ToT) algorithm for strategic look ahead planning and resource aware navigation, and (iii) a Dialectical Diagnostic Verification procedure utilizing "Angel-Devil" adversarial debates to resolve complex evidence conflicts. Evaluated on two real world benchmarks, MIMIC-IV-Ext Cardiac Disease and MIMIC-IV-Ext CDM, DxChain achieves state-of-the-art performances in both diagnostic accuracy and logical consistency, offering a modular and reliable architecture for next-generation clinical AI. The code is at https://anonymous.4open.science/r/Dx-Chain.

cs.AI↗
arXiv2026

Electrically steered conduction topologies and period-doubling phase dynamics in VO2

The insulator-to-metal transition (IMT) in strongly correlated materials, such as vanadium dioxide (VO2), offers a transformative platform for next-generation adaptive electronics and neuromorphic computing. However, harnessing this non-equilibrium phase transition for deterministic device operation is fundamentally hindered by the inability to disentangle electric-field effects from Joule heating, owing to a lack of operando techniques capable of resolving phase dynamics at nanoscale spatial and sub-nanosecond temporal scales. Here, using a newly developed electrical-pulse-pump ultrafast transmission electron microscope (E-UTEM), we directly visualize the multi-scale electro-thermo-mechanical dynamics of the IMT in suspended VO2 devices. Our results reveal that electric-field-induced Poole-Frenkel (PF) emission, localized by patterned oxygen vacancies, plays a decisive role in redistributing the internal electric field to trigger a deterministic Mott transition. The extreme non-linearity of this PF effect enables the formation of dynamically reconfigurable connectivity topologies that bypass conventional thermal limits. Furthermore, we observe that the coupling of thermal and elastic energies governs a discrete domain evolution, characterized by step-wise and period-doubling configurational resets, which is a hallmark of non-equilibrium phase dynamics in constrained geometries. By integrating experimental imaging with phase-field simulations, we establish a comprehensive framework for the electrically-driven IMT and predict sub-100-ps switching kinetics. These findings provide a fundamental basis for the rational design of ultrafast, low-energy functional devices through nanoscale defect and strain engineering in correlated systems.

cond-mat.mtrl-sci↗
arXiv2026

HiPrune: Hierarchical Attention for Efficient Token Pruning in Vision-Language Models

Vision-Language Models (VLMs) encode images and videos into abundant tokens, which contain substantial redundancy and computation cost. While visual token pruning mitigates the issue, most existing methods lack insight into the intrinsic property of the vision encoder itself. In this work, we dive into the vision encoder and prove that the middle layers pay more attention to the main objects of the image qualitatively and quantitatively, while the deep layers to tokens with rich global information. Utilizing this Hierarchical attention pattern, we propose HiPrune, a training-free and model-agnostic token Pruning method. HiPrune identifies three types of visual tokens according to their attention in different phases of the vision encoder, which preserves different levels of information. By coupling with the similarity of text tokens, we propose a prompt-aware variance, HiPrune++, which further improves instruction following performance under a very low token budget. Extensive experiments across four representative VLMs show that HiPrune achieves up to 99.3% of task accuracy with only 1/3 of the tokens, while reducing inference FLOPs by 58.7%. HiPrune++ maintains up to 99.7% accuracy with 2/9 tokens, highlighting robustness under high-resolution. Our code is available at https://github.com/Danielement321/HiPrune.

cs.CV↗
arXiv2026

Long time smooth solutions of 3D cubic quasilinear wave systems with small weakly decaying initial data

For the 3D cubic quasilinear wave system $\square_{c_i} u^i=G^i(u,\partial u,\partial^2u)=\displaystyle\sum_{\substack{0\le|α|,|β|,|γ|\le1 \\ 1\le j,k,l \le m}}g_{αβγ}^{ijkl}\partial^αu^j\partial^βu^k\partial^γu^l$, it is well known that global solution $u$ exists when the small smooth initial data $(u,\partial_tu)|_{t=0}$ $=(u_0(x), u_1(x))$ are compactly supported or decay rapidly at spatial infinity. However, when $(u_0, u_1)\in (H^{s+1}, H^s)$ with $s>\frac{5}{2}$ are small, it remains unknown whether $u$ exists globally or not. In this paper, we show that if $\|u_{0}\|_{H^{N+1}}+\|u_{1}\|_{H^N}\le\varepsilon$ ($N\ge 6$) is small, then the almost global solution $u$ exists in $[0, T_{\varepsilon}]$ with $T_{\varepsilon}\ge e^{C\varepsilon^{-1}}$ for the general $G(u,\partial u,\partial^2u)$ depending on $u$ and $T_{\varepsilon}\ge e^{C\varepsilon^{-2}}$ for the nonlinearity $G(\partial u,\partial^2u)$ independent of $u$, respectively. In addition, if $\displaystyle\sum_{|a|\le 5}\|\langle x\rangle^μ\partial^a_x(u _0,u_1)\|_{L^2}\le\varepsilon$ holds for any fixed constant $μ\in (0,1)$, then the solution $u$ exists globally and meanwhile the scattering property of $u$ is derived. Our main ingredients consist in establishing a series of new weighted $L^\infty-L^2$ estimates and Strichartz estimates based on the strong Huygens' principle for 3D linear wave equations.

math.AP↗
arXiv2026

Beyond Text Prompts: Precise Concept Erasure through Text-Image Collaboration

Text-to-image generative models have achieved impressive fidelity and diversity, but can inadvertently produce unsafe or undesirable content due to implicit biases embedded in large-scale training datasets. Existing concept erasure methods, whether text-only or image-assisted, face trade-offs: textual approaches often fail to fully suppress concepts, while naive image-guided methods risk over-erasing unrelated content. We propose TICoE, a text-image Collaborative Erasing framework that achieves precise and faithful concept removal through a continuous convex concept manifold and hierarchical visual representation learning. TICoE precisely removes target concepts while preserving unrelated semantic and visual content. To objectively assess the quality of erasure, we further introduce a fidelity-oriented evaluation strategy that measures post-erasure usability. Experiments on multiple benchmarks show that TICoE surpasses prior methods in concept removal precision and content fidelity, enabling safer, more controllable text-to-image generation. Our code is available at https://github.com/OpenAscent-L/TICoE.git

cs.CV↗
arXiv2026

Evo-MedAgent: Beyond One-Shot Diagnosis with Agents That Remember, Reflect, and Improve

Tool-augmented large language model (LLM) agents can orchestrate specialist classifiers, segmentation models, and visual question-answering modules to interpret chest X-rays. However, these agents still solve each case in isolation: they fail to accumulate experience across cases, correct recurrent reasoning mistakes, or adapt their tool-use behavior without expensive reinforcement learning. While a radiologist naturally improves with every case, current agents remain static. In this work, we propose Evo-MedAgent, a self-evolving memory module that equips a medical agent with the capacity for inter-case learning at test time. Our memory comprises three complementary stores: (1)~\emph{Retrospective Clinical Episodes} that retrieve problem-solving experiences from similar past cases, (2)~an \emph{Adaptive Procedural Heuristics} bank curating priority-tagged diagnostic rules that evolves via reflection, much like a physician refining their internal criteria, and (3)~a \emph{Tool Reliability Controller} that tracks per-tool trustworthiness. On ChestAgentBench, Evo-MedAgent raises multiple-choice question (MCQ) accuracy from 0.68 to 0.79 on GPT-5-mini, and from 0.76 to 0.87 on Gemini-3 Flash. With a strong base model, evolving memory improves performance more effectively than orchestrating external tools on qualitative diagnostic tasks. Because Evo-MedAgent requires no training, its per-case overhead is bounded by one additional retrieval pass and a single reflection call, making it deployable on top of any frozen model.

cs.AI↗
arXiv2026

Scaling Exposes the Trigger: Input-Level Backdoor Detection in Text-to-Image Diffusion Models via Cross-Attention Scaling

Text-to-image (T2I) diffusion models have achieved remarkable success in image synthesis, but their reliance on large-scale data and open ecosystems introduces serious backdoor security risks. Existing defenses, particularly input-level methods, are more practical for deployment but often rely on observable anomalies that become unreliable under stealthy, semantics-preserving trigger designs. As modern backdoor attacks increasingly embed triggers into natural inputs, these methods degrade substantially, raising a critical question: can more stable, implicit, and trigger-agnostic differences between benign and backdoor inputs be exploited for detection? In this work, we address this challenge from an active probing perspective. We introduce controlled scaling perturbations on cross-attention and uncover a novel phenomenon termed Cross-Attention Scaling Response Divergence (CSRD), where benign and backdoor inputs exhibit systematically different response evolution patterns across denoising steps. Building on this insight, we propose SET, an input-level backdoor detection framework that constructs response-offset features under multi-scale perturbations and learns a compact benign response space from a small set of clean samples. Detection is then performed by measuring deviations from this learned space, without requiring prior knowledge of the attack or access to model training. Extensive experiments demonstrate that SET consistently outperforms existing baselines across diverse attack methods, trigger types, and model settings, with particularly strong gains under stealthy implicit-trigger scenarios. Overall, SET improves AUROC by 9.1% and ACC by 6.5% over the best baseline, highlighting its effectiveness and robustness for practical deployment.

cs.CR↗