Search arXiv⌕ Search

arXiv subjects

Yuan Gao

Publications and source records attributed to Yuan Gao.

At least 145 records · Page 8Linked to original sources
arXiv2026

DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing

Diffusion models have made significant progress in both text-to-image (T2I) generation and text-guided image editing. However, these models are typically built with billions of parameters, leading to high latency and increased deployment challenges. While on-device diffusion models improve efficiency, they largely focus on T2I generation and lack support for image editing. In this paper, we propose DreamLite, a compact unified on-device diffusion model (0.39B) that supports both T2I generation and text-guided image editing within a single network. DreamLite is built on a pruned mobile U-Net backbone and unifies conditioning through in-context spatial concatenation in the latent space. It concatenates images horizontally as input, using a (target | blank) configuration for generation tasks and (target | source) for editing tasks. To stabilize the training of this compact model, we introduce a task-progressive joint pretraining strategy that sequentially targets T2I, editing, and joint tasks. After high-quality SFT and reinforcement learning, DreamLite achieves GenEval (0.72) for image generation and ImgEdit (4.11) for image editing, outperforming existing on-device models and remaining competitive with several server-side models. By employing step distillation, we further reduce denoising processing to just 4 steps, enabling our DreamLite could generate or edit a 1024 x 1024 image in less than 1s on a Xiaomi 14 smartphone. To the best of our knowledge, DreamLite is the first unified on-device diffusion model that supports both image generation and image editing.

cs.CV↗
arXiv2026

AnalogAgent: Self-Improving Analog Circuit Design Automation with LLM Agents

Recent advances in large language models (LLMs) suggest strong potential for automating analog circuit design. Yet most LLM-based approaches rely on a single-model loop of generation, diagnosis, and correction, which favors succinct summaries over domain-specific insight and suffers from context attrition that erases critical technical details. To address these limitations, we propose AnalogAgent, a training-free agentic framework that integrates an LLM-based multi-agent system (MAS) with self-evolving memory (SEM) for analog circuit design automation. AnalogAgent coordinates a Code Generator, Design Optimizer, and Knowledge Curator to distill execution feedback into an adaptive playbook in SEM and retrieve targeted guidance for subsequent generation, enabling cross-task transfer without additional expert feedback, databases, or libraries. Across established benchmarks, AnalogAgent achieves 92% Pass@1 with Gemini and 97.4% Pass@1 with GPT-5. Moreover, with compact models (e.g., Qwen-8B), it yields a +48.8% average Pass@1 gain across tasks and reaches 72.1% Pass@1 overall, indicating that AnalogAgent substantially strengthens open-weight models for high-quality analog circuit design automation.

cs.AI↗
arXiv2026

RKKY-dipolar Interactions and 3D Spin Supersolid on Stacked Triangular Lattice

Inspired by the recent discovery of metallic spin supersolidity and its giant magnetocaloric effect in the rare-earth alloy EuCo$_2$Al$_9$ [Nature 651, 61 (2026)], we perform a combined study through electronic structure analysis, effective spin model, and Monte Carlo simulations on a stacked triangular lattice, and reveal a novel mechanism for the emergence of 3D spin supersolid in a metallic antiferromagnet. From first-principles inputs, we derive a minimal spin model on a stacked triangular lattice (STL), which arises from the interplay between Ruderman-Kittel-Kasuya-Yosida (RKKY) and dipolar interactions and accurately reproduces the experimental thermodynamics. Based on the STL model, we identify a ground state that simultaneously breaks discrete lattice translational symmetry and continuous spin-rotational symmetry -- the hallmark of a spin supersolid. Furthermore, we present the field-temperature phase diagram of the 3D STL model and discuss the various magnetic phases and associated phase transitions. Under zero field, the spin supersolid Y order establishes in two steps: an upper transition at $T_{N1}$, where an emergent U(1) symmetry appears and the system enters a fluctuating collinear regime, followed by a lower transition at $T_{N2}$ into the spin supersolid Y phase. In contrast, the supersolid V phase undergoes a single phase transition at $T_N^V$. Our results not only provide a comprehensive theoretical understanding of the metallic spin supersolid reported for EuCo$_2$Al$_9$ but also pave the way for further experimental investigations into its supersolid transitions and universality class.

cond-mat.str-el↗
arXiv2026

Baillon-Bruck-Reich revisited: divergent-series parameters and strong convergence in the linear case

The Krasnoselskii-Mann iteration is an important algorithm in optimization and variational analysis for finding fixed points of nonexpansive mappings. In the general case, it produces a sequence converging \emph{weakly} to a fixed point provided the parameter sequence satisfies a divergent-series condition. In this paper, we show that \emph{strong} convergence holds provided the underlying nonexpansive mapping is \emph{linear}. This improves on a celebrated result by Baillon, Bruck, and Reich from 1978, where the parameter sequence was assumed to be constant as well as on recent work where the parameters were bounded away from $0$ and $1$.

math.OC↗
arXiv2026

Imaging foundation model for universal enhancement of non-ideal measurement CT

Non-ideal measurement computed tomography (NICT) employs suboptimal imaging protocols to expand CT applications. However, the resulting trade-offs degrade image quality, limiting clinical acceptability. Although deep learning methods have been used to enhance NICT images, their reliance on large training datasets and limited generalizability across diverse settings hinder practical use. We propose the multi-scale integrated Transformer AMPlifier (TAMP), the first imaging foundation model for universal NICT enhancement. Pre-trained on 10.8 million physics-driven simulated NICT images, TAMP generalizes effectively across various NICT settings, defect degrees, and body regions. Moreover, a parameter-efficient fine-tuning strategy enables TAMP to adapt to specific clinical scenarios using only few slices. Extensive experiments, including radiologists and real-world validations, demonstrate that TAMP consistently improves image quality and clinical acceptability, underscoring its significant potential to advance CT imaging and broaden NICT applications in clinical practice.

eess.IV↗
arXiv2026

Timestep-Aware Block Masking for Efficient Diffusion Model Inference

Diffusion Probabilistic Models (DPMs) have achieved great success in image generation but suffer from high inference latency due to their iterative denoising nature. Motivated by the evolving feature dynamics across the denoising trajectory, we propose a novel framework to optimize the computational graph of pre-trained DPMs on a per-timestep basis. By learning timestep-specific masks, our method dynamically determines which blocks to execute or bypass through feature reuse at each inference stage. Unlike global optimization methods that incur prohibitive memory costs via full-chain backpropagation, our method optimizes masks for each timestep independently, ensuring a memory-efficient training process. To guide this process, we introduce a timestep-aware loss scaling mechanism that prioritizes feature fidelity during sensitive denoising phases, complemented by a knowledge-guided mask rectification strategy to prune redundant spatial-temporal dependencies. Our approach is architecture-agnostic and demonstrates significant efficiency gains across a broad spectrum of models, including DDPM, LDM, DiT, and PixArt. Experimental results show that by treating the denoising process as a sequence of optimized computational paths, our method achieves a superior balance between sampling speed and generative quality. Our code will be released.

cs.CV↗
arXiv2026

OMNIFLOW: A Physics-Grounded Multimodal Agent for Generalized Scientific Reasoning

Large Language Models (LLMs) have demonstrated exceptional logical reasoning capabilities but frequently struggle with the continuous spatiotemporal dynamics governed by Partial Differential Equations (PDEs), often resulting in non-physical hallucinations. Existing approaches typically resort to costly, domain-specific fine-tuning, which severely limits cross-domain generalization and interpretability. To bridge this gap, we propose OMNIFLOW, a neuro-symbolic architecture designed to ground frozen multimodal LLMs in fundamental physical laws without requiring domain-specific parameter updates. OMNIFLOW introduces a novel \textit{Semantic-Symbolic Alignment} mechanism that projects high-dimensional flow tensors into topological linguistic descriptors, enabling the model to perceive physical structures rather than raw pixel values. Furthermore, we construct a Physics-Guided Chain-of-Thought (PG-CoT) workflow that orchestrates reasoning through dynamic constraint injection (e.g., mass conservation) and iterative reflexive verification. We evaluate OMNIFLOW on a comprehensive benchmark spanning microscopic turbulence, theoretical Navier-Stokes equations, and macroscopic global weather forecasting. Empirical results demonstrate that OMNIFLOW significantly outperforms traditional deep learning baselines in zero-shot generalization and few-shot adaptation tasks. Crucially, it offers transparent, physically consistent reasoning reports, marking a paradigm shift from black-box fitting to interpretable scientific reasoning.

cs.LG↗
arXiv2026

Speech-Worthy Alignment for Japanese SpeechLLMs via Direct Preference Optimization

SpeechLLMs typically combine ASR-trained encoders with text-based LLM backbones, leading them to inherit written-style output patterns unsuitable for text-to-speech synthesis. This mismatch is particularly pronounced in Japanese, where spoken and written registers differ substantially in politeness markers, sentence-final particles, and syntactic complexity. We propose a preference-based alignment approach to adapt Japanese SpeechLLMs for speech-worthy outputs: text that is concise, conversational, and readily synthesized as natural speech. To rigorously evaluate this task, we introduce SpokenElyza, a benchmark for Japanese speech-worthiness derived from ELYZA-tasks-100 with auditory verification by native experts. Experiments show that our approach achieves substantial improvement on SpokenElyza while largely preserving performance on the original written-style evaluation. We will release SpokenElyza to support future research on Japanese spoken dialog systems.

cs.SD↗
arXiv2026

Distilling LLM Semantic Priors into Encoder-Only Multi-Talker ASR with Talker-Count Routing

Large language models (LLMs) provide strong semantic priors that can improve multi-talker automatic speech recognition (MT-ASR), but using an LLM as an autoregressive decoder is computationally expensive and remains fragile under heavy overlap. In this paper, we propose an encoder-only MT-ASR framework that adapts an LLM to multi-talker conditioning and distills its semantic guidance into the encoder during training, while retaining fast CTC-style decoding at inference. Our model employs a post-encoder separator with serialized CTC to produce talker-ordered transcripts, and leverages an adapted LLM-based SOT objective as a multi-talker-aware teacher signal to explicitly regularize mixed-speech representations. To further support variable numbers of talkers, we introduce a Talker-Count Head that predicts the talker count and dynamically selects the appropriate decoding branch. Experiments on LibriMix show that the proposed encoder-only model achieves comparable performance to LLM-based systems in the two-talker condition, while delivering significant improvements in the three-talker condition with significant small RTF.

cs.SD↗
arXiv2026

Low-loss phase-change material based programmable mode converter for photonic computing

Phase-change materials (PCMs)-based integrated photonic memory offers a viable pathway for the development of neuromorphic computing chip. The sizable optical contrast in the telecom band between amorphous and crystalline phases of PCM, in particular, Ge2Sb2Te5 (GST), is used for multilevel programming. However, the high extinction coefficient k of crystalline GST leads to high optical loss, posing a serious challenge for scaling up the device array for practical use. In this work, we focus on the atomic understanding and application of the so-called low-loss PCM, Sb2Se3, through multiscale simulations. First, we elucidate the bonding origin of the wavelength dependent optical properties of amorphous and crystalline Sb2Se3 via ab initio calculations. Given the suppressed k in the telecom band, we design a programable mode converter (PMC) waveguide device that utilizes only the contrast in refractive index n between amorphous and crystalline Sb2Se3 to encode multiple optical levels per waveguide device. The finite-difference time-domain simulations show that a single PMC device can achieve 5-bit programming precision (32 levels) via direct laser writing, and the photonic tensor core formed by the PMC array could possibly be scaled to 128*128. Finally, a thorough comparison between low-loss PCM and conventional PCM is provided.

cond-mat.mtrl-sci↗
arXiv2026

Paralinguistic Emotion-Aware Validation Timing Detection in Japanese Empathetic Spoken Dialogue

Emotional Validation is a psychotherapy communication technique that involves recognizing, understanding, and explicitly acknowledging another person's feelings and actions, which strengthens alliance and reduces negative affect. To maximize the emotional support provided by validation, it is crucial to deliver it with appropriate timing and frequency. This study investigates validation timing detection from the speech perspective. Leveraging both paralinguistic and emotional information, we propose a paralinguistic- and emotion-aware model for validation timing detection without relying on textual context. Specifically, we first conduct continued self-supervised training and fine-tuning on different HuBERT backbones to obtain (i) a paralinguistics-aware Self-Supervised Learning (SSL) encoder and (ii) a multi-task speech emotion classification encoder. We then fuse these encoders and further fine-tune the combined model on the downstream validation timing detection task. Experimental evaluations on the TUT Emotional Storytelling Corpus (TESC) compare multiple models, fusion mechanisms, and training strategies, and demonstrate that the proposed approach achieves significant improvements over conventional speech baselines. Our results indicate that non-linguistic speech cues, when integrated with affect-related representations, carry sufficient signal to decide when validation should be expressed, offering a speech-first pathway toward more empathetic human-robot interaction.

cs.SD↗
arXiv2026

StyleVLA: Driving Style-Aware Vision Language Action Model for Autonomous Driving

Vision Language Models (VLMs) bridge visual perception and linguistic reasoning. In Autonomous Driving (AD), this synergy has enabled Vision Language Action (VLA) models, which translate high-level multimodal understanding into driving behaviors, typically represented as future trajectories. However, existing VLA models mainly generate generic collision-free trajectories. Beyond collision avoidance, adapting to diverse driving styles (e.g., sporty, comfortable) is essential for personalized driving. Moreover, many methods treat trajectory generation as naive token prediction, which can produce kinematically infeasible actions. To address these limitations, we present StyleVLA, a physics-informed VLA framework for generating diverse and physically plausible driving behaviors. We introduce a hybrid loss that combines a kinematic consistency constraint with a continuous regression head to improve trajectory feasibility. To train StyleVLA, built on Qwen3-VL-4B, we construct a large-scale instruction dataset with over 1.2k scenarios, 76k Bird's Eye View (BEV) samples, and 42k First Person View (FPV) samples, with ground-truth trajectories for five driving styles and natural-language instructions. Experiments show that our 4B-parameter StyleVLA significantly outperforms proprietary models (e.g., Gemini-3-Pro) and state-of-the-art VLA models. Using a composite driving score measuring success rate, physical feasibility, and style adherence, StyleVLA achieves 0.55 on BEV and 0.51 on FPV, versus 0.32 and 0.35 for Gemini-3-Pro. These results show that a specialized, physics-informed, lightweight model can surpass closed-source models on domain-specific tasks.

cs.RO↗
arXiv2026

TumorChain: Interleaved Multimodal Chain-of-Thought Reasoning for Traceable Clinical Tumor Analysis

Accurate tumor analysis is central to clinical radiology and precision oncology, where early detection, reliable lesion characterization, and pathology-level risk assessment guide diagnosis and treatment planning. Chain-of-Thought (CoT) reasoning is particularly important in this setting because it enables step-by-step interpretation from imaging findings to clinical impressions and pathology conclusions, improving traceability and reducing diagnostic errors. Here, we target the clinical tumor analysis task and build a large-scale benchmark that operationalizes a multimodal reasoning pipeline, spanning findings, impressions, and pathology predictions. We curate TumorCoT, a large-scale dataset of 1.5M CoT-labeled VQA instructions paired with 3D CT scans, with step-aligned rationales and cross-modal alignments along the trajectory from findings to impression to pathology, enabling evaluation of both answer accuracy and reasoning consistency. We further propose TumorChain, a multimodal interleaved reasoning framework that tightly couples 3D imaging encoders, clinical text understanding, and organ-level vision-language alignment. Through cross-modal alignment and iterative interleaved causal reasoning, TumorChain grounds visual evidence, aggregates conclusions, and issues pathology predictions after multiple rounds of self-refinement, improving traceability and reducing hallucination risk. Experiments show consistent improvements over strong baselines in lesion detection, impression generation, and pathology classification, and demonstrate strong generalization on the DeepTumorVQA benchmark. These results highlight the potential of multimodal reasoning for reliable and interpretable tumor analysis in clinical practice. Detailed information about our project can be found on our project homepage at https://github.com/ZJU4HealthCare/TumorChain.

cs.CV↗
arXiv2026

Passive All-Optical Nonlinear Neuron Activation via PPLN Nanophotonic Waveguides

Artificial intelligence (AI) is transforming modern life, yet the growing scale of AI applications places mounting demands on computational resources, raising sustainability concerns. Photonic integrated circuits (PICs) offer a promising alternative, enabling massive parallelism, low latency, and reduced electrical overhead, particularly excelling in high-throughput linear operations. However, passive and fully optical nonlinear activation functions with equally superb performance remain rare, posing a critical bottleneck in realizing all-optical neural networks in PICs. Here, we demonstrate a compact and integrated all-optical nonlinear activation method, experimentally realized through strong second-order optical nonlinearities in periodically poled lithium niobate (PPLN) nanophotonic waveguides, achieving 80% absolute conversion efficiency. This activation exhibits a sigmoid-like, wavelength-selective response with femtosecond-scale dynamics and light-speed processing, operating passively without external control and auxiliary signals. We validate its feasibility for neural inference by cascading the PPLN-driven activations with a linear silicon PIC, demonstrating all-optical nonlinear neuron expressivity. Moreover, combining the measured nonlinearity with linear operations calculated by the PIC, we show that PPLN-activated multi-layer optical neural networks can achieve performance on par with digital implementations in real-world tasks, including airfoil regression and medical image classification. These results pave the way toward scalable, high-speed, and fully integrated all-optical neural networks for next-generation photonic AI hardware.

physics.optics↗
arXiv2026

Pailitao-VL: Unified Embedding and Reranker for Real-Time Multi-Modal Industrial Search

In this work, we presented Pailitao-VL, a comprehensive multi-modal retrieval system engineered for high-precision, real-time industrial search. We here address three critical challenges in the current SOTA solution: insufficient retrieval granularity, vulnerability to environmental noise, and prohibitive efficiency-performance gap. Our primary contribution lies in two fundamental paradigm shifts. First, we transitioned the embedding paradigm from traditional contrastive learning to an absolute ID-recognition task. Through anchoring instances to a globally consistent latent space defined by billions of semantic prototypes, we successfully overcome the stochasticity and granularity bottlenecks inherent in existing embedding solutions. Second, we evolved the generative reranker from isolated pointwise evaluation to the compare-and-calibrate listwise policy. By synergizing chunk-based comparative reasoning with calibrated absolute relevance scoring, the system achieves nuanced discriminative resolution while circumventing the prohibitive latency typically associated with conventional reranking methods. Extensive offline benchmarks and online A/B tests on Alibaba e-commerce platform confirm that Pailitao-VL achieves state-of-the-art performance and delivers substantial business impact. This work demonstrates a robust and scalable path for deploying advanced MLLM-based retrieval architectures in demanding, large-scale production environments.

cs.IR↗
arXiv2026

DuaLip-GPU Technical Report

Large-scale linear programs (LPs) arise in many decision systems, including ranking, allocation, and matching problems that must be solved repeatedly at massive scale. Prior work such as ECLIPSE and LinkedIn's open-source DuaLip showed that ridge-regularized dual ascent with first-order methods can scale to these settings. However, the original implementation was tightly coupled to a small number of schemas and built on a CPU-centric Scala/Spark stack, limiting extensibility and preventing effective use of modern accelerators. We present a redesigned solver architecture that decouples problem specification from the optimization engine and targets GPU execution. The system uses an operator-centric programming model in which LP formulations are expressed through composable primitives for dual objective evaluation and blockwise projection operators for decomposable constraint families. This design allows new formulations to be added locally while reusing a shared optimization loop, diagnostics, and distributed infrastructure. To realize the available parallelism, we develop GPU execution techniques tailored to sparse matching constraints, including constraint-aligned sparse layouts, batched projection kernels, and a distributed design that communicates only dual variables. Further, we improve the underlying ridge-regularized dual ascent method with Jacobi-style row normalization, primal scaling, and a continuation scheme for the regularization parameter. On extreme-scale matching workloads, the GPU implementation achieves at least a 10x wall-clock speedup over the prior distributed CPU DuaLip solver under matched stopping criteria, while maintaining convergence guarantees.

cs.DC↗
arXiv2026

MOON: Generative MLLM-based Multimodal Representation Learning for E-commerce Product Understanding

With the rapid advancement of e-commerce, exploring general representations rather than task-specific ones has attracted increasing research attention. For product understanding, although existing discriminative dual-flow architectures drive progress in this field, they inherently struggle to model the many-to-one alignment between multiple images and texts of products. Therefore, we argue that generative Multimodal Large Language Models (MLLMs) hold significant potential for improving product representation learning. Nevertheless, achieving this goal still remains non-trivial due to several key challenges: the lack of multimodal and aspect-aware modeling modules in typical LLMs; the common presence of background noise in product images; and the absence of a standard benchmark for evaluation. To address these issues, we propose the first generative MLLM-based model named MOON for product representation learning. Our method (1) employs a guided Mixture-of-Experts (MoE) module for targeted modeling of multimodal and aspect-specific product content; (2) effectively detects core semantic regions in product images to mitigate the distraction and interference caused by background noise; and (3) introduces the specialized negative sampling strategy to increase the difficulty and diversity of negative samples. In addition, we release a large-scale multimodal benchmark MBE for various product understanding tasks. Experimentally, our model demonstrates competitive zero-shot performance on both our benchmark and the public dataset, showcasing strong generalization across various downstream tasks, including cross-modal retrieval, product classification, and attribute prediction. Furthermore, the case study and visualization illustrate the effectiveness of MOON for product understanding. The data of our MBE benchmark is given in https://huggingface.co/datasets/Daoze/MM-Bench-E-Commerce.

cs.CV↗
arXiv2026

Producing $Λ(1405)$ and $Λ(1520)$ in $π^-p$ reaction to explore their inner structures

In this work, the production mechanisms of the hyperon resonances $Λ(1405)$ and $Λ(1520)$ in the $π^- p$ scattering are investigated within an effective Lagrangian approach incorporating Regge trajectories. By including contributions from $t$-channel $K^*$ and $u$-channel $Σ$ exchanges, we perform global fits to the total and differential cross sections for $π^{-} p \rightarrow KΛ(1405)$ and $π^{-} p \rightarrow KΛ(1520)$. The results show good agreement with available experimental data. For the total cross section of $Λ(1405)$ production, the $u$-channel contribution is dominant, whereas the $t$-channel contribution plays the primary role in $Λ(1520)$ production. Furthermore, the differential cross sections of the two processes exhibit distinctly different shapes, reflecting their distinct underlying reaction mechanisms. An analysis based on the constituent counting rule indicates that $Λ(1520)$ is consistent with a conventional three-quark configuration, while $Λ(1405)$ shows a clear deviation, suggesting a more exotic structure. Owing to the large branching ratio of $Λ^* \to πΣ$, the Dalitz process $π^{-} p \rightarrow K Λ^{*} \rightarrow K πΣ$ is also calculated. Our results demonstrate that reconstructing $Λ^*$ via the $KπΣ$ final state is experimentally feasible. This study provides important theoretical insights into the production dynamics of these hyperon resonances, and suggests future high-precision measurements of the $t$-distribution at large momentum transfer at facilities such as AMBER, J-PARC, HIKE, and HIAF, which can further clarify their reaction mechanisms and structural properties.

nucl-th↗