arXiv subjects

Hongyu Li

Publications and source records attributed to Hongyu Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

Visualizing the interplay of dual electronic nematicities in kagome superconductors

Kagome superconductor AV$_3$Sb$_5$ (A stands for K, Rb, and Cs) hosts a wealth of intertwined electronic orders driven by geometric frustration and electron correlations. Among them, the breaking of rotational and/or time-reversal symmetry, observed within the triple-$Q$ charge density wave (CDW) phase yet exhibiting a more complex temperature dependence, remains a central puzzle. Here, by using scanning tunneling microscopy to study the electronic structures of CsV$_3$Sb$_5$ as a function of temperature and Ti doping, we disentangle the interrelation between two distinct nematic order parameters, one associated with the CDW and the other manifested as $C_2$ distortion of the V-$d_{x^{2}-y^{2}}$ Fermi pockets without breaking transition symmetry. The latter persists to high doping levels and high temperatures where the long-range CDW is fully suppressed. Moreover, its nematic director is oriented in a lattice direction distinct from that of the CDW-induced nematicity at intermediate doping, and eventually aligns with the strong nematic CDW order in the pristine compound where the quasiparticles of vanadium orbitals become coherent below a lower characteristic temperature. These observations, combined with Ginzburg-Landau analysis, reveal a rich interplay between two nematic orders that can be assigned to distinct kagome-lattice orbitals. Our results shed new light on the enigmatic intertwined orders in this family and establish a rare material platform in which dual nematic orders coexist and couple to give rise to unusual correlated phenomena.

cond-mat.supr-con↗

arXiv2026

LongCat-Next: Lexicalizing Modalities as Discrete Tokens

The prevailing Next-Token Prediction (NTP) paradigm has driven the success of large language models through discrete autoregressive modeling. However, contemporary multimodal systems remain language-centric, often treating non-linguistic modalities as external attachments, leading to fragmented architectures and suboptimal integration. To transcend this limitation, we introduce Discrete Native Autoregressive (DiNA), a unified framework that represents multimodal information within a shared discrete space, enabling a consistent and principled autoregressive modeling across modalities. A key innovation is the Discrete Native Any-resolution Visual Transformer (dNaViT), which performs tokenization and de-tokenization at arbitrary resolutions, transforming continuous visual signals into hierarchical discrete tokens. Building on this foundation, we develop LongCat-Next, a native multimodal model that processes text, vision, and audio under a single autoregressive objective with minimal modality-specific design. As an industrial-strength foundation model, it excels at seeing, painting, and talking within a single framework, achieving strong performance across a wide range of multimodal benchmarks. In particular, LongCat-Next addresses the long-standing performance ceiling of discrete vision modeling on understanding tasks and provides a unified approach to effectively reconcile the conflict between understanding and generation. As an attempt toward native multimodality, we open-source the LongCat-Next and its tokenizers, hoping to foster further research and development in the community. GitHub: https://github.com/meituan-longcat/LongCat-Next

cs.CV↗

arXiv2026

RPiAE: A Representation-Pivoted Autoencoder Enhancing Both Image Generation and Editing

Diffusion models have become the dominant paradigm for image generation and editing, with latent diffusion models shifting denoising to a compact latent space for efficiency and scalability. Recent attempts to leverage pretrained visual representation models as tokenizer priors either align diffusion features to representation features or directly reuse representation encoders as frozen tokenizers. Although such approaches can improve generation metrics, they often suffer from limited reconstruction fidelity due to frozen encoders, which in turn degrades editing quality, as well as overly high-dimensional latents that make diffusion modeling difficult. To address these limitations, We propose Representation-Pivoted AutoEncoder, a representation-based tokenizer that improves both generation and editing. We introduce Representation-Pivot Regularization, a training strategy that enables a representation-initialized encoder to be fine-tuned for reconstruction while preserving the semantic structure of the pretrained representation space, followed by a variational bridge which compress latent space into a compact one for better diffusion modeling. We adopt an objective-decoupled stage-wise training strategy that sequentially optimizes generative tractability and reconstruction-fidelity objectives. Together, these components yield a tokenizer that preserves strong semantics, reconstructs faithfully, and produces latents with reduced diffusion modeling complexity. Experiments demonstrate that RPiAE outperforms other visual tokenizers on text-to-image generation and image editing, while delivering the best reconstruction fidelity among representation-based tokenizers.

cs.CV↗

arXiv2026

Mass Production of 2023 KMTNet Microlensing Planets. II: Two Planets and A Brown Dwarf

To expand the homogeneous microlensing planetary sample of the Korea Microlensing Telescope Network (KMTNet), we investigate six planetary candidates identified by the AnomalyFinder search in the 2023 prime-field data, namely KMT-2023-BLG-1592, OGLE-2023-BLG-0766, KMT-2023-BLG-0332, KMT-2023-BLG-0486, KMT-2023-BLG-0792, and OGLE-2023-BLG-1043. Light-curve modeling indicates that the first two events have planetary mass ratios of $\log q \sim -3.0$ and $-2.6$, while the third exhibits a brown dwarf mass ratio of $\log q \sim -1.4$. The remaining three events show the well-known degeneracy between the binary-lens single-source (2L1S) and single-lens binary-source (1L2S) interpretations. A Bayesian analysis yields companion masses of about 0.6 and 1.2 Jupiter masses for the two planetary systems, likely orbiting beyond the snow lines of M- or K-dwarf hosts. A review of the KMTNet planetary sample shows that candidates discovered by AnomalyFinder are significantly more likely to exhibit the 2L1S/1L2S degeneracy, consistent with the tendency of AnomalyFinder to detect subtler planetary signals.

astro-ph.EP↗

arXiv2026

UMO: Unified In-Context Learning Unlocks Motion Foundation Model Priors

Large-scale foundation models (LFMs) have recently made impressive progress in text-to-motion generation by learning strong generative priors from massive 3D human motion datasets and paired text descriptions. However, how to effectively and efficiently leverage such single-purpose motion LFMs, i.e., text-to-motion synthesis, in more diverse cross-modal and in-context motion generation downstream tasks remains largely unclear. Prior work typically adapts pretrained generative priors to individual downstream tasks in a task-specific manner. In contrast, our goal is to unlock such priors to support a broad spectrum of downstream motion generation tasks within a single unified framework. To bridge this gap, we present UMO, a simple yet general unified formulation that casts diverse downstream tasks into compositions of atomic per-frame operations, enabling in-context adaptation to unlock the generative priors of pretrained DiT-based motion LFMs. Specifically, UMO introduces three learnable frame-level meta-operation embeddings to specify per-frame intent and employs lightweight temporal fusion to inject in-context cues into the pretrained backbone, with negligible runtime overhead compared to the base model. With this design, UMO finetunes the pretrained model, originally limited to text-to-motion generation, to support diverse previously unsupported tasks, including temporal inpainting, text-guided motion editing, text-serialized geometric constraints, and multi-identity reaction generation. Experiments demonstrate that UMO consistently outperforms task-specific and training-free baselines across a wide range of benchmarks, despite using a single unified model. Code and model will be publicly available. Project Page: https://oliver-cong02.github.io/UMO.github.io/

cs.CV↗

arXiv2026

Two Low Mass-Ratio Microlensing Planets and Two Types of Central-Resonant Degeneracy

We present observations and analysis of two low planet/host mass-ratio ($q$) microlensing planets discovered in high-magnification events. KMT-2025-BLG-0811Lb has $q \sim 4.5 \times 10^{-5}$, and a Bayesian analysis favors a super-Earth/mini-Neptune orbiting an M- or K-dwarf host at a projected separation of $\sim 3$ au. KMT-2025-BLG-0912Lb has $q = 2.6 \times 10^{-4}$ and likely hosts a super-Earth/mini-Neptune around either a low-mass M dwarf or a brown dwarf at $\sim 1$ au. Even with an observing cadence of $Γ> 30~{\rm hr}^{-1}$ during the planetary signal, KMT-2025-BLG-0811 still exhibits the "central-resonant" degeneracy. Reviewing nine such events, we find that the "central-resonant" degeneracy can be divided into two distinct types that occupy separate regions in the plane of $q$ and normalized source radius ($ρ$). Type~I events have similar $q$ but substantially different $ρ$ and are more difficult to resolve from the light curves. For Type~II events, the "resonant" solutions have relatively lower $q$ and larger $ρ$. Our review provides guidance for searching for the alternative solution once one solution has been identified.

astro-ph.EP↗

arXiv2026

Antenna Coding Optimization for Pixel Antenna Empowered Wireless Communication Using Deep Learning with Heterogeneous Multi-Head Selection

Pixel antenna is a promising antenna technology that enables flexible adjustment of radiation characteristics and enhancement of wireless systems through antenna coding. This work proposes a novel deep learning-based antenna coding optimization algorithm. Specifically, the proposed algorithm is supported by a heterogeneous multi-head selection mechanism, whose main idea is to train multiple neural networks based on various coding schemes and select the one that leads to the best system performance. Unlike traditional heuristic searching-based algorithms that require high computational complexity to achieve satisfactory performance, the proposed data-driven deep learning approach can achieve 98\% of the performance achieved by the searching-based algorithms with significantly reduced computational complexity. Results demonstrate that in pixel antenna empowered single-input single-output systems, the proposed algorithm achieves a computational speed 81 times faster than the searching-based algorithm. For more complex pixel antenna empowered multiple-input multiple-output systems, the computational speed is 297 times faster than the existing searching-based algorithm. Benefiting from the high performance and low computational complexity, this algorithm demonstrates the significant potential of pixel antennas as a novel and practical technology to enhance wireless systems.

cs.IT↗

arXiv2026

Characterizing Robustness of Strategies to Novelty in Zero-Sum Open Worlds

In open-world environments, artificial agents must often contend with novel conditions that deviate from their training or design assumptions. This paper studies the robustness of fixed-strategy agents to such novelty within the setting of two-player zero-sum games. We present a general framework for characterizing the impact of environmental novelties, such as changes in payoff structure or action constraints, on agent performance in two distinct domains: Iterated Prisoner's Dilemma (IPD) and heads-up Texas Hold'em Poker. Novelty is operationalized as a perturbation of the game's rules or scoring mechanics, while agent behavior remains fixed. To measure the effects, we introduce two metrics: per-agent robustness, quantifying the relative performance shift of each strategy across novelties, and global impact, summarizing the population-wide disruption caused by a novelty. Our experiments, comprising 30 IPD agents across 20 payoff matrix novelties and 10 Poker agents across 5 rule-based novelties, reveal systematic patterns in robustness and highlight certain novelties that induce severe destabilization. The results offer insights into agent generalizability under perturbation and provide a quantitative basis for designing safer and more resilient autonomous systems in adversarial and dynamic environments.

cs.GT↗

arXiv2026

LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion

The safety mechanisms of large language models (LLMs) exhibit notable fragility, as even fine-tuning on datasets without harmful content may still undermine their safety capabilities. Meanwhile, existing safety alignment methods predominantly rely on the fine-tuning process, which inadvertently leads to the increased complexity and computational resources required. To address these issues, we introduce LSSF, a novel safety re-alignment framework with \underline{L}ow-Rank \underline{S}afety \underline{S}ubspace \underline{F}usion. Our proposed method exploits the low-rank characteristics of safety information in LLMs by constructing a low-rank projection matrix to extract the principal components of safety vectors. Notably, this projection matrix represents the low-rank safety subspace of the LLMs, which we have observed to remain stable during fine-tuning process and is isolated from the model's general capabilities. These principal components are used to effectively restore safety alignment when combined with fine-tuned LLMs through linear arithmetic. Additionally, to account for the varying encoding densities of safety information across different layers of LLMs, we propose a novel metric called safety singular value entropy. This metric quantifies the encoding density and allows for the dynamic computation of the safety-critical rank for each safety vector. Extensive experiments demonstrate that our proposed post-hoc alignment method can effectively restore the safety alignment of fine-tuned models with minimal impact on their performance in downstream tasks.

cs.CY↗

arXiv2026

KMT-2025-BLG-1616Lb: First Microlensing Bound Planet From DREAMS

We present observations and analysis of the bound planetary microlensing event KMT-2025-BLG-1616. The planetary signal was captured by the Korea Microlensing Telescope Network (KMTNet) and the DECam Rogue Earths and Mars Survey (DREAMS). DREAMS's minute-cadence observations break the central/resonant degeneracy in the binary-lens models. The color of the faint source star ($I=22$) is measured from the DREAMS's $r - z$ color. The planetary system has a planet-host mass ratio of $q \sim 5 \times 10^{-4}$. A Bayesian analysis yields a host-star mass of $\sim 0.3\,M_\odot$, a planetary mass of $\sim 40\,M_{\oplus}$, a projected planet-host separation of $\sim 1.6~\mathrm{au}$, and a lens distance of $\sim 7.5~\mathrm{kpc}$. Based on the photometric precision achieved by DREAMS for this event, we simulate free-floating planet (FFP) detections and find that DREAMS is sensitive to Mars-mass FFPs in the Galactic bulge and Moon-mass FFPs in the Galactic disk.

astro-ph.EP↗

arXiv2026

Antenna Coding Optimization for Pixel Antenna Empowered MIMO Wireless Power Transfer

We investigate antenna coding utilizing pixel antennas as a new degree of freedom for enhancing multiple-input multiple-output (MIMO) wireless power transfer (WPT) systems. The objective is to enhance the output direct current (DC) power under RF combining and DC combining schemes by jointly exploiting gains from antenna coding, beamforming, and rectenna nonlinearity. We first propose the MIMO WPT system model with binary and continuous antenna coding using the beamspace channel model and formulate the joint antenna coding and beamforming optimization using a nonlinear rectenna model. We propose two efficient closed-form successive convex approximation algorithms to efficiently optimize the beamforming. To further reduce the computational complexity, we propose codebook-based antenna coding designs for output DC power maximization based on K-means clustering. Results show that the proposed pixel antenna empowered MIMO WPT system with binary antenna coding increases output DC power by more than 15 dB compared with conventional systems with fixed antenna configuration. With continuous antenna coding, the performance improves another 6 dB. Moreover, the proposed codebook design outperforms previous designs by up to 40% and shows good performance with reduced computational complexity. Overall, the significant improvement in output DC power verifies the potential of leveraging antenna coding utilizing pixel antennas to enhance WPT systems.

eess.SP↗

arXiv2025

A Tutorial on Beyond-Diagonal Reconfigurable Intelligent Surfaces: Modeling, Architectures, System Design and Optimization, and Applications

Written by its inventors, this first tutorial on Beyond-Diagonal Reconfigurable Intelligent Surfaces (BD-RISs) provides the readers with the basics and fundamental tools necessary to appreciate, understand, and contribute to this emerging and disruptive technology. Conventional (Diagonal) RISs (D-RISs) are characterized by a diagonal scattering matrix $\mathbfΘ$ such that the wave manipulation flexibility of D-RIS is extremely limited. In contrast, BD-RIS refers to a novel and general framework for RIS where its scattering matrix is not limited to be diagonal (hence, the ``beyond-diagonal'' terminology) and consequently, all entries of $\mathbfΘ$ can potentially help shaping waves for much higher manipulation flexibility. This physically means that BD-RIS can artificially engineer and reconfigure coupling across elements of the surface thanks to inter-element reconfigurable components which allow waves absorbed by one element to flow through other elements. Consequently, BD-RIS opens the door to more general and versatile intelligent surfaces that subsumes existing RIS architectures as special cases. In this tutorial, we share all the secret sauce to model, design, and optimize BD-RIS and make BD-RIS transformative in many different applications. Topics discussed include physics-consistent and multi-port network-aided modeling; transmitting, reflecting, hybrid, and multi-sector mode analysis; reciprocal and non-reciprocal architecture designs and optimal performance-complexity Pareto frontier of BD-RIS; signal processing, optimization, and channel estimation for BD-RIS; hardware impairments (discrete-value impedance and admittance, lossy interconnections and components, wideband effects, mutual coupling) of BD-RIS; benefits and applications of BD-RIS in communications, sensing, power transfer.

eess.SP↗

arXiv2025

Antenna Coding Optimization Based on Pixel Antennas for MIMO Wireless Power Transfer with DC Combining

This paper investigates antenna coding based on pixel antennas as a new degree of freedom for enhancing multiple-input multiple-output (MIMO) wireless power transfer (WPT) systems. Antenna coding is closely related to the Fluid Antenna System (FAS) concept and further generalizes the radiation pattern reconfigurability. We first introduce a beamspace channel model to demonstrate reconfigurable radiation patterns enabled by antenna coders. By jointly optimizing the antenna coding and transmit beamforming with perfect channel state information (CSI), we exploit gains from antenna coding, transmit beamforming, and rectenna nonlinearity to maximize the output DC power. We adopt an alternating optimization approach with the quasi-Newton method and Successive Exhaustive Boolean Optimization (SEBO) method with warm-start to handle the transmit beamforming design and antenna coding design respectively. Finally, simulation results show that the proposed MIMO WPT system with pixel antennas achieves up to 15 dB gain in average output DC power compared with a conventional system with fixed antenna configuration, highlighting the potential of pixel antennas for boosting the WPT efficiency.

eess.SP↗

arXiv2025

Antenna Coding Design for Multi-User Transmissions Using Pixel Antennas

This work investigates exploiting the potential of pixel antennas, which are a reconfigurable antenna technology that can flexibly adjust the antenna characteristics through antenna coding, in multi-user transmissions. To that end, we propose a multi-user multi-input single-output (MISO) pixel antenna system, which deploys the pixel antenna at users, and develop the system model including pixel antenna with antenna coding and multi-user beamspace channels. Aiming at maximizing the sum rate performance, we first propose an algorithm to alternatively design the precoding at the transmitter and the antenna coding at users, which explores the performance boundary for the proposed multi-user MISO pixel antenna system. To reduce the computational complexity, we propose a codebook-based antenna coding design algorithm, where the antenna coder is online optimized from an offline codebook. To further enhance the computation efficiency, we propose a hierarchical codebook-based antenna coding design that uses a multi-layer hierarchical search to achieve a better performance-complexity trade-off. Simulation results show that, adopting the proposed algorithms, the multi-user MISO pixel antenna system can always outperform conventional multi-user MISO systems with fixed antennas. More importantly, results validate that the proposed (hierarchical) codebook-based algorithms can significantly reduce the computational complexity while maintaining a satisfactory sum rate performance.

eess.SP↗

arXiv2025

OpenSubject: Leveraging Video-Derived Identity and Diversity Priors for Subject-driven Image Generation and Manipulation

Despite the promising progress in subject-driven image generation, current models often deviate from the reference identities and struggle in complex scenes with multiple subjects. To address this challenge, we introduce OpenSubject, a video-derived large-scale corpus with 2.5M samples and 4.35M images for subject-driven generation and manipulation. The dataset is built with a four-stage pipeline that exploits cross-frame identity priors. (i) Video Curation. We apply resolution and aesthetic filtering to obtain high-quality clips. (ii) Cross-Frame Subject Mining and Pairing. We utilize vision-language model (VLM)-based category consensus, local grounding, and diversity-aware pairing to select image pairs. (iii) Identity-Preserving Reference Image Synthesis. We introduce segmentation map-guided outpainting to synthesize the input images for subject-driven generation and box-guided inpainting to generate input images for subject-driven manipulation, together with geometry-aware augmentations and irregular boundary erosion. (iv) Verification and Captioning. We utilize a VLM to validate synthesized samples, re-synthesize failed samples based on stage (iii), and then construct short and long captions. In addition, we introduce a benchmark covering subject-driven generation and manipulation, and then evaluate identity fidelity, prompt adherence, manipulation consistency, and background consistency with a VLM judge. Extensive experiments show that training with OpenSubject improves generation and manipulation performance, particularly in complex scenes.

cs.CV↗

arXiv2025

EditThinker: Unlocking Iterative Reasoning for Any Image Editor

Instruction-based image editing has emerged as a prominent research area, which, benefiting from image generation foundation models, have achieved high aesthetic quality, making instruction-following capability the primary challenge. Existing approaches improve instruction adherence via supervised or reinforcement learning, yet single-turn success rates remain limited due to inherent stochasticity and a lack of deliberation. In this work, we propose a deliberative editing framework to 'think' while they edit, which simulates the human cognitive loop by iteratively executing a Think-while-Edit cycle: Critiquing results and Refining instructions , followed by Repeating the generation until satisfactory. Specifically, we train a single MLLM, EditThinker, to act as the reasoning engine of this framework, which jointly produce the critique score, reasoning process, and refined instructions. We employ reinforcement learning to align the EditThinker's thinking with its editing, thereby generating more targeted instruction improvements. Extensive experiments on four benchmarks demonstrate that our approach significantly improves the instruction-following capability of any image editing model by a large margin. We will release our data construction framework, datasets, and models to benefit the community.

cs.CV↗

arXiv2025

Shadows in the Code: Exploring the Risks and Defenses of LLM-based Multi-Agent Software Development Systems

The rapid advancement of Large Language Model (LLM)-driven multi-agent systems has significantly streamlined software developing tasks, enabling users with little technical expertise to develop executable applications. While these systems democratize software creation through natural language requirements, they introduce significant security risks that remain largely unexplored. We identify two risky scenarios: Malicious User with Benign Agents (MU-BA) and Benign User with Malicious Agents (BU-MA). We introduce the Implicit Malicious Behavior Injection Attack (IMBIA), demonstrating how multi-agent systems can be manipulated to generate software with concealed malicious capabilities beneath seemingly benign applications, and propose Adv-IMBIA as a defense mechanism. Evaluations across ChatDev, MetaGPT, and AgentVerse frameworks reveal varying vulnerability patterns, with IMBIA achieving attack success rates of 93%, 45%, and 71% in MU-BA scenarios, and 71%, 84%, and 45% in BU-MA scenarios. Our defense mechanism reduced attack success rates significantly, particularly in the MU-BA scenario. Further analysis reveals that compromised agents in the coding and testing phases pose significantly greater security risks, while also identifying critical agents that require protection against malicious user exploitation. Our findings highlight the urgent need for robust security measures in multi-agent software development systems and provide practical guidelines for implementing targeted, resource-efficient defensive strategies.

cs.CR↗

arXiv2025

Thinking-while-Generating: Interleaving Textual Reasoning throughout Visual Generation

Recent advances in visual generation have increasingly explored the integration of reasoning capabilities. They incorporate textual reasoning, i.e., think, either before (as pre-planning) or after (as post-refinement) the generation process, yet they lack on-the-fly multimodal interaction during the generation itself. In this preliminary study, we introduce Thinking-while-Generating (TwiG), the first interleaved framework that enables co-evolving textual reasoning throughout the visual generation process. As visual content is progressively generating, textual reasoning is interleaved to both guide upcoming local regions and reflect on previously synthesized ones. This dynamic interplay produces more context-aware and semantically rich visual outputs. To unveil the potential of this framework, we investigate three candidate strategies, zero-shot prompting, supervised fine-tuning (SFT) on our curated TwiG-50K dataset, and reinforcement learning (RL) via a customized TwiG-GRPO strategy, each offering unique insights into the dynamics of interleaved reasoning. We hope this work inspires further research into interleaving textual reasoning for enhanced visual generation. Code will be released at: https://github.com/ZiyuGuo99/Thinking-while-Generating.

cs.CV↗