Yanxin Zhang

Enyi Shi, Pengyang Shao, Yanxin Zhang et al.

Robust safety of vision-language large models (VLLMs) under joint multilingual and multimodal inputs remains underexplored. Existing benchmarks are typically multilingual but text-only, or multimodal but monolingual. Recent multilingual multimodal red-teaming efforts render harmful prompts into images, yet rely heavily on typography-style visuals and lack semantically grounded image-text pairs, limiting coverage of realistic cross-modal interactions. We introduce Lingua-SafetyBench, a benchmark of 100,440 harmful image-text pairs across 10 languages, explicitly partitioned into image-dominant and text-dominant subsets to disentangle risk sources. Evaluating 11 open-source VLLMs reveals a consistent asymmetry: image-dominant risks yield higher ASR in high-resource languages, while text-dominant risks are more severe in non-high-resource languages. A controlled study on the Qwen series shows that scaling and version upgrades reduce Attack Success Rate (ASR) overall but disproportionately benefit HRLs, widening the gap between HRLs and Non-HRLs under text-dominant risks. This underscores the necessity of language- and modality-aware safety alignment beyond mere scaling.To facilitate reproducibility and future research, we will publicly release our benchmark, model checkpoints, and source code.The code and dataset will be available at https://github.com/zsxr15/Lingua-SafetyBench.Warning: this paper contains examples with unsafe content.

Shuibai Zhang, Caspian Zhuang, Chihan Cui et al.

Diffusion language models (DLMs) enable parallel, non-autoregressive text generation, yet existing DLM mixture-of-experts (MoE) models inherit token-choice (TC) routing from autoregressive systems, leading to load imbalance and rigid computation allocation. We show that expert-choice (EC) routing is a better fit for DLMs: it provides deterministic load balancing by design, yielding higher throughput and faster convergence than TC. Building on the property that EC capacity is externally controllable, we introduce timestep-dependent expert capacity, which varies expert allocation according to the denoising step. We find that allocating more capacity to low-mask-ratio steps consistently achieves the best performance under matched FLOPs, and provide a mechanistic explanation: tokens in low-mask-ratio contexts exhibit an order-of-magnitude higher learning efficiency, so concentrating compute on these steps yields the largest marginal return. Finally, we show that existing pretrained TC DLMs can be retrofitted to EC by replacing only the router, achieving faster convergence and improved accuracy across diverse downstream tasks. Together, these results establish EC routing as a superior paradigm for DLM MoE models and demonstrate that computation in DLMs can be treated as an adaptive policy rather than a fixed architectural constant. Code is available at https://github.com/zhangshuibai/EC-DLM.

Huatuan Sun, Yunshan Ma, Changguang Wu et al.

Frozen Large Video Language Models (LVLMs) are increasingly employed in micro-video recommendation due to their strong multimodal understanding. However, their integration lacks systematic empirical evaluation: practitioners typically deploy LVLMs as fixed black-box feature extractors without systematically comparing alternative representation strategies. To address this gap, we present the first systematic empirical study along two key design dimensions: (i) integration strategies with ID embeddings, specifically replacement versus fusion, and (ii) feature extraction paradigms, comparing LVLM-generated captions with intermediate decoder hidden states. Extensive experiments on representative LVLMs reveal three key principles: (1) intermediate hidden states consistently outperform caption-based representations, as natural-language summarization inevitably discards fine-grained visual semantics crucial for recommendation; (2) ID embeddings capture irreplaceable collaborative signals, rendering fusion strictly superior to replacement; and (3) the effectiveness of intermediate decoder features varies significantly across layers. Guided by these insights, we propose the Dual Feature Fusion (DFF) Framework, a lightweight and plug-and-play approach that adaptively fuses multi-layer representations from frozen LVLMs with item ID embeddings. DFF achieves state-of-the-art performance on two real-world micro-video recommendation benchmarks, consistently outperforming strong baselines and providing a principled approach to integrating off-the-shelf large vision-language models into micro-video recommender systems.

Yanxin Zhang, Ian McDougall, Junnan Li et al.

For over a decade, processor design has focused on implementing sophisticated policies for various components of the out-of-order pipeline, including cache replacement and prefetching. The prevailing design philosophy has been to build processors with a single, static selection of policies across these different mechanisms. This paper investigates a fundamental question: do different workloads, or even different execution phases within the same workload, benefit from different policy combinations? We present a comprehensive analysis exploring whether a hypothetical processor capable of dynamically selecting from multiple policies could significantly outperform traditional static-policy processors. Using ChampSim-based simulation across 49 benchmarks segmented into 490 execution phases of 20M instructions each, we evaluate performance across multiple policy combinations for cache replacement and prefetching. Our findings reveal that significant performance headroom exists: the best static policy achieves optimal performance for only 19.18\% of execution phases and incurs a mean IPC loss of 1.54\% compared to an oracle. Moreover, 85 phases (17.35\%), spanning 14 of the 49 applications, exhibit more than 2.5\% IPC loss relative to the oracle. Furthermore, we demonstrate that a processor capable of dynamically switching between two carefully chosen policies can achieve a 13.6$\times$ reduction in mean IPC loss (from 1.54\% to 0.11\%) and match oracle performance 52.65\% of the time. These results suggest that dynamic policy selection represents a promising avenue for unlocking single-thread performance improvements that have become increasingly difficult to achieve.

Yixin Xiang, Yunshan Ma, Xiaoyu Du et al.

Document Question Answering (DQA) involves generating answers from a document based on a user's query, representing a key task in document understanding. This task requires interpreting visual layouts, which has prompted recent studies to adopt multimodal Retrieval-Augmented Generation (RAG) that processes page images for answer generation. However, in multimodal RAG, visual DQA struggles to utilize a large number of images effectively, as the retrieval stage often retains only a few candidate pages (e.g., Top-4), causing informative but less visually salient content to be overlooked in favor of common yet low-information pages. To address this issue, we propose a Multi-Armed Bandit-based DQA framework (MAB-DQA) to explicitly model the varying importance of multiple implicit aspects in a query. Specifically, MAB-DQA decomposes a query into aspect-aware subqueries and retrieves an aspect-specific candidate set for each. It treats each subquery as an arm and uses preliminary reasoning results from a small number of representative pages as reward signals to estimate aspect utility. Guided by an exploration-exploitation policy, MAB-DQA dynamically reallocates retrieval budgets toward high-value aspects. With the most informative pages and their correlations, MAB-DQA generates the expected results. On four benchmarks, MAB-DQA shows an average improvement of 5%-18% over the state-of-the-art method, consistently enhancing document understanding. Code at https://github.com/ElephantOH/MAB-DQA.

Sen Fang, Yalin Feng, Yanxin Zhang et al.

In this paper, we propose a Rectified Flow Auto Coder (RAC) inspired by Rectified Flow to replace the traditional VAE: 1. It achieves multi-step decoding by applying the decoder to flow timesteps. Its decoding path is straight and correctable, enabling step-by-step refinement. 2. The model inherently supports bidirectional inference, where the decoder serves as the encoder through time reversal (hence Coder rather than encoder or decoder), reducing parameter count by nearly 41%. 3. This generative decoding method improves generation quality since the model can correct latent variables along the path, partially addressing the reconstruction--generation gap. Experiments show that RAC surpasses SOTA VAEs in both reconstruction and generation with approximately 70% lower computational cost.

Sen Fang, Yalin Feng, Hongbin Zhong et al.

Sign Language Production (SLP) is the process of converting the complex input text into a real video. Most previous works focused on the Text2Gloss, Gloss2Pose, Pose2Vid stages, and some concentrated on Prompt2Gloss and Text2Avatar stages. However, this field has made slow progress due to the inaccuracy of text conversion, pose generation, and the rendering of poses into real human videos in these stages, resulting in gradually accumulating errors. Therefore, in this paper, we streamline the traditional redundant structure, simplify and optimize the task objective, and design a new sign language generative model called Stable Signer. It redefines the SLP task as a hierarchical generation end-to-end task that only includes text understanding (Prompt2Gloss, Text2Gloss) and Pose2Vid, and executes text understanding through our proposed new Sign Language Understanding Linker called SLUL, and generates hand gestures through the named SLP-MoE hand gesture rendering expert block to end-to-end generate high-quality and multi-style sign language videos. SLUL is trained using the newly developed Semantic-Aware Gloss Masking Loss (SAGM Loss). Its performance has improved by 48.6% compared to the current SOTA generation methods.

Sen Fang, Hongbin Zhong, Yanxin Zhang et al.

Existing large-scale sign language resources typically provide supervision only at the level of raw video-text alignment and are often produced in laboratory settings. While such resources are important for semantic understanding, they do not directly provide a unified interface for open-world recognition and translation, or for modern pose-driven sign language video generation frameworks: 1. RGB-based pretrained recognition models depend heavily on fixed backgrounds or clothing conditions during recording, and are less robust in open-world settings than style-agnostic pose-processing models. 2. Recent pose-guided image/video generation models mostly use a unified keypoint representation such as DWPose as their control interface. At present, the sign language field still lacks a data resource that can directly interface with this modern pose-native paradigm while also targeting real-world open scenarios. We present SignVerse-2M, a large-scale multilingual pose-native dataset for sign language pose modeling and evaluation. Built from publicly available multilingual sign language video resources, it applies DWPose in a unified preprocessing pipeline to convert raw videos into 2D pose sequences that can be used directly for modeling, resulting in a consolidated corpus of about two million clips covering more than 25 sign languages. Unlike many laboratory datasets, this resource preserves the recording conditions and speaker diversity of real-world videos while reducing appearance variation through a unified pose representation. Toward this goal, we further provide the data construction pipeline, task definitions, and a simple SignDW Transformer baseline, demonstrating the feasibility of this resource for multilingual pose-space modeling and its compatibility with modern pose-driven pipelines, while discussing the evaluation claims it can support as well as its current limitations.

Hongyang Yang, Yanxin Zhang, Yang She et al.

Housing selection is a high-stakes and largely irreversible decision problem. We study housing consultation as a decision-support interface for housing selection. Existing housing platforms and many LLM-based assistants often reduce this process to ranking or recommendation, resulting in opaque reasoning, brittle multi-constraint handling, and limited guarantees on factuality. We present HabitatAgent, the first LLM-powered multi-agent architecture for end-to-end housing consultation. HabitatAgent comprises four specialized agent roles: Memory, Retrieval, Generation, and Validation. The Memory Agent maintains multi-layer user memory through internal stages for constraint extraction, memory fusion, and verification-gated updates; the Retrieval Agent performs hybrid vector--graph retrieval (GraphRAG); the Generation Agent produces evidence-referenced recommendations and explanations; and the Validation Agent applies multi-tier verification and targeted remediation. Together, these agents provide an auditable and reliable workflow for end-to-end housing consultation. We evaluate HabitatAgent on 100 real user consultation scenarios (300 multi-turn question--answer pairs) under an end-to-end correctness protocol. A strong single-stage baseline (Dense+Rerank) achieves 75% accuracy, while HabitatAgent reaches 95%.

Sen Fang, Hongbin Zhong, Yalin Feng et al.

New technologies such as Rectified Flow and Flow Matching have significantly improved the performance of generative models in the past two years, especially in terms of control accuracy, generation quality, and generation efficiency. However, due to some differences in its theory, design, and existing diffusion models, the existing acceleration methods cannot be directly applied to the Rectified Flow model. In this article, we have comprehensively implemented an overall acceleration pipeline from the aspects of theory, design, and reasoning strategies. This pipeline uses new methods such as batch processing with a new velocity field, vectorization of heterogeneous time-step batch processing, and dynamic TensorRT compilation for the new methods to comprehensively accelerate related models based on flow models. Currently, the existing public methods usually achieve an acceleration of 18%, while experiments have proved that our new method can accelerate the 512*512 image generation speed to up to 611%, which is far beyond the current non-generalized acceleration methods.

Yongzhi Li, Saining Zhang, Yibing Chen et al.

Personalized image generation aims to faithfully preserve a reference subject's identity while adapting to diverse text prompts. Existing optimization-based methods ensure high fidelity but are computationally expensive, while learning-based approaches offer efficiency at the cost of entangled representations influenced by nuisance factors. We introduce SpotDiff, a novel learning-based method that extracts subject-specific features by spotting and disentangling interference. Leveraging a pre-trained CLIP image encoder and specialized expert networks for pose and background, SpotDiff isolates subject identity through orthogonality constraints in the feature space. To enable principled training, we introduce SpotDiff10k, a curated dataset with consistent pose and background variations. Experiments demonstrate that SpotDiff achieves more robust subject preservation and controllable editing than prior methods, while attaining competitive performance with only 10k training samples.

Shayne Wadle, Yanxin Zhang, Vikas Singh et al.

The evaluation of new microprocessor designs is constrained by slow, cycle-accurate simulators that rely on unrepresentative benchmark traces. This paper introduces a novel deep learning framework for high-fidelity, ``in-the-wild'' simulation on production hardware. Our core contribution is a DL model trained on microarchitecture-independent features to predict cycle-level performance for hypothetical processor designs. This unique approach allows the model to be deployed on existing silicon to evaluate future hardware. We propose a complete system featuring a lightweight hardware trace collector and a principled sampling strategy to minimize user impact. This system achieves a simulation speed of 5 MIPS on a commodity GPU, imposing a mere 0.1% performance overhead. Furthermore, our co-designed Neutrino on-chip accelerator improves performance by 85x over the GPU. We demonstrate that this framework enables accurate performance analysis and large-scale hardware A/B testing on a massive scale using real-world applications.

Zeyi Kang, Liang He, Yanxin Zhang et al.

Multimodal semantic learning plays a critical role in embodied intelligence, especially when robots perceive their surroundings, understand human instructions, and make intelligent decisions. However, the field faces technical challenges such as effective fusion of heterogeneous data and computational efficiency in resource-constrained environments. To address these challenges, this study proposes the lightweight LCMF cascaded attention framework, introducing a multi-level cross-modal parameter sharing mechanism into the Mamba module. By integrating the advantages of Cross-Attention and Selective parameter-sharing State Space Models (SSMs), the framework achieves efficient fusion of heterogeneous modalities and semantic complementary alignment. Experimental results show that LCMF surpasses existing multimodal baselines with an accuracy of 74.29% in VQA tasks and achieves competitive mid-tier performance within the distribution cluster of Large Language Model Agents (LLM Agents) in EQA video tasks. Its lightweight design achieves a 4.35-fold reduction in FLOPs relative to the average of comparable baselines while using only 166.51M parameters (image-text) and 219M parameters (video-text), providing an efficient solution for Human-Robot Interaction (HRI) applications in resource-constrained scenarios with strong multimodal decision generalization capabilities.

Longtao Zhu, Hongyu Yang, Ge Song et al.

Current flight procedure design methods heavily rely on human-led design process, which is not only low auto-mation but also suffer from complex algorithm modelling and poor generalization. To address these challenges, this paper proposes an agent-driven flight procedure design method based on large language model, named Au-toFPDesigner, which utilizes multi-agent collaboration to complete procedure design. The method enables end-to-end automated design of performance-based navigation (PBN) procedures. In this process, the user input the design requirements in natural language, AutoFPDesigner models the flight procedure design by loading the design speci-fications and utilizing tool libraries complete the design. AutoFPDesigner allows users to oversee and seamlessly participate in the design process. Experimental results show that AutoFPDesigner ensures nearly 100% safety in the designed flight procedures and achieves 75% task completion rate, with good adaptability across different design tasks. AutoFPDesigner introduces a new paradigm for flight procedure design and represents a key step towards the automation of this process. Keywords: Flight Procedure Design; Large Language Model; Performance-Based Navigation (PBN); Multi Agent;

Haoran Zhang, Chenkun Yin, Yanxin Zhang et al.

In this paper, two novel practical methods of Reinforcement Learning from Demonstration (RLfD) are developed and applied to automatic berthing control systems for Unmanned Surface Vessel. A new expert data generation method, called Model Predictive Based Expert (MPBE) which combines Model Predictive Control and Deep Deterministic Policy Gradient, is developed to provide high quality supervision data for RLfD algorithms. A straightforward RLfD method, model predictive Deep Deterministic Policy Gradient (MP-DDPG), is firstly introduced by replacing the RL agent with MPBE to directly interact with the environment. Then distribution mismatch problem is analyzed for MP-DDPG, and two techniques that alleviate distribution mismatch are proposed. Furthermore, another novel RLfD algorithm based on the MP-DDPG, called Self-Guided Actor-Critic (SGAC) is present, which can effectively leverage MPBE by continuously querying it to generate high quality expert data online. The distribution mismatch problem leading to unstable learning process is addressed by SGAC in a DAgger manner. In addition, theoretical analysis is given to prove that SGAC algorithm can converge with guaranteed monotonic improvement. Simulation results verify the effectiveness of MP-DDPG and SGAC to accomplish the ship berthing control task, and show advantages of SGAC comparing with other typical reinforcement learning algorithms and MP-DDPG.