Dan Zhao

Wenhao Li, Xueying Jiang, Quanhao Qian et al.

Despite rapid progress, most existing vision-language models (VLMs) built from 2D visual inputs often struggle when handling various 3D tasks that require fine-grained spatial understanding and reasoning. To bridge this gap, we present VLM-IE3D, a unified framework that enhances the 3D spatial awareness of VLMs by equipping them with both implicit and explicit 3D geometries learned from RGB videos. Our VLM-IE3D introduces Implicit Geometry Tokens (IGTs) that capture high-level geometric priors from input videos, as well as complementary Explicit Geometry Tokens (EGTs) that encode detailed geometric structures from reconstructed 3D attributes. On top of that, VLM-IE3D comes with a 3D-aware adapter that effectively fuses the two types of geometric representations with 2D visual cues. This RGB-only design injects strong 3D inductive biases for fine-grained spatial understanding and reasoning without requiring any additional 3D inputs. Extensive experiments show that VLM-IE3D achieves superior performance consistently across various 3D tasks including 3D video detection, 3D visual grounding, 3D dense captioning, and spatial reasoning. Code and models are available at https://github.com/Vegetebird/VLM-IE3D.

Shixin Guo, Ming Zhong, Xuhong Zhang et al.

Generating high-quality adversarial texts with low query budgets remains a challenging problem in the hard-label scenario. Most existing approaches rely on greedy algorithms, where one position in the text is selected for substitution, followed by the substitutions of other positions. This local search approach may fail to discover high-quality adversarial examples and often leads to excessive query costs. Ideally, an optimal adversarial sample would consider all possible position combinations in the text, but exhaustive search is computationally impractical. To address this challenge, we propose a sampling-based method called LBA, which constructs an approximate distribution of high-quality adversarial examples by integrating both prior and posterior knowledge, and utilizes this distribution for sampling. As sampling progresses, posterior knowledge updates the approximate distribution, which in turn guides more effective sampling. Extensive experiments on six language models, ranging from small-scale to large-scale architectures across four datasets, demonstrate that LBA significantly outperforms state-of-the-art baselines on all evaluation metrics. Additionally, LLM-based assessment indicates that LBA generates more semantically preserved and comprehensible adversarial texts.

Guoyang Zhao, Quanhao Qian, Gongjie Zhang et al.

Proprioception is fundamental to robotic manipulation, yet standard fusion methods often treat it as an isolated vector lacking explicit alignment with visual tokens. Without a direct correspondence between 3D kinematics and 2D feature maps, manipulation policies struggle to ground the robot's state within the scene, frequently underperforming even vision-only baselines. To address this, we introduce GeoProp, a lightweight, plug-and-play adapter that aligns proprioception with vision through explicit geometric grounding and spatial feature sampling. GeoProp projects the robot state onto the image plane to sample localized visual features, constructing a grounded state token. It then injects state-derived spatial priors into the corresponding visual features via FiLM modulation. To capture motion intent, GeoProp further samples features at a short-horizon predicted coordinate derived from recent kinematics, providing look-ahead visual context. Across 67 tasks, GeoProp improves Diffusion Policy by 8.7% on 63 simulation tasks and pi_0 by 4.0% on the RoboTwin subset, and yields a 10.6% average gain across both policy families in the real world, while adding only 2-3% to the parameter count. These results demonstrate that GeoProp is a simple yet high-impact inductive bias for generalist embodied policies. Project page: https://alibaba-damo-academy.github.io/GeoProp/.

Xin Cheng, Xingkai Yu, Chenze Shao et al.

Speculative decoding accelerates Large Language Model (LLM) inference by decoupling draft generation from target verification. While recent parallel drafters efficiently propose long token sequences in a single forward pass, they suffer from rapid acceptance decay due to a lack of inter-token dependencies. Furthermore, indiscriminately verifying these extended blocks wastes critical batch capacity on tokens with high rejection risks, severely degrading throughput in high-concurrency serving systems. We introduce DSpark, a speculative decoding framework that unifies high-throughput parallel generation with adaptive, load-aware verification. To maintain draft quality, DSpark utilizes a semi-autoregressive architecture, coupling a parallel backbone with a lightweight sequential module, to introduce intra-block dependency modeling and mitigate suffix decay. To optimize system efficiency, DSpark employs confidence-scheduled verification, dynamically tailoring the verification length for each request based on estimated prefix survival probabilities and engine-specific throughput profiles. On offline benchmarks across diverse domains, DSpark substantially improves the accepted length over state-of-the-art autoregressive and parallel drafters. When deployed within the DeepSeek-V4 serving system under live user traffic, DSpark successfully mitigates verification waste. Compared to the established production baseline (MTP-1), DSpark accelerates per-user generation speeds by 60 to 85 percent at matched throughput levels. More importantly, by preventing severe throughput degradation under strict interactivity constraints, it enables performance tiers that were previously unattainable, shifting the Pareto frontier of our serving system.

Shuai Tian, Yupeng Zheng, Yuhang Zheng et al.

Contact-rich manipulation requires policies to react to local deformation, pressure, slip, and friction, yet these cues are temporally sparse and often invisible in visual observations. Existing visual-tactile policies usually feed tactile observations directly into action prediction, but rarely model tactile deformation dynamics during action generation. In this paper, we introduce VT-WAM, a Visual-Tactile World Action Model that jointly learns future visual prediction, tactile deformation prediction, and action prediction within a unified flow matching framework. In particular, VT-WAM introduces (1) Asymmetric Mixture-of-Transformers (MoT) attention to bridge a first-frame visual anchor with temporal tactile dynamics, and (2) contact-gated Action-Visual-Tactile Attention Guidance (AVTAG) to encourage action queries to rely on tactile evidence during contact phases. Across six real-world contact-rich manipulation tasks, VT-WAM achieves a 71.67% average success rate, outperforming Fast-WAM by 26.67% and OmniVTLA by 35.84%. Ablations demonstrate that modeling tactile deformation dynamics and guiding contact-phase tactile attention are both important for contact-rich tasks. Project website: https://vt-wam.github.io/.

Dankai Liao, Tianyi Zhang, Yufeng Wu et al.

Whole-slide image (WSI) diagnosis requires identifying diagnostically relevant regions, examining them across magnifications, and integrating multi-scale evidence. However, most existing pathology benchmarks evaluate models on pre-cropped patches or pre-extracted slide features, leaving their ability to acquire evidence directly from gigapixel WSIs largely untested. We introduce PathAgentBench, a benchmark for evaluating evidence-seeking vision-language models (VLMs) across four complementary capabilities: image-to-text matching for evidence interpretation, text-to-image retrieval for evidence verification, diagnostic-region localization for evidence acquisition, and multi-scale reasoning for evidence integration. The benchmark is organized as a diagnostic tree that links nested regions across magnifications with scale-specific findings and path-level diagnoses. It contains 1,822 TCGA WSIs and 17,135 diagnostic paths annotated by ten board-certified pathologists. An additional private cohort of 190 breast cancer WSIs with detailed annotations is used to evaluate autonomous whole-slide exploration. We evaluate 20 general-purpose, medical, and pathology-specialized models. Leading open-weight models achieve over 93% accuracy in multi-scale reasoning and over 50% accuracy in both cross-modal matching tasks. In contrast, diagnostic-region localization remains challenging: the best text-guided mean intersection-over-union is below 0.09, underperforming a simple center-based heuristic. During autonomous exploration, the unconditional hit rate decreases from 0.522 at low magnification to 0.185 at intermediate magnification and 0.020 at high magnification. These results reveal a pronounced gap between reasoning over curated evidence and acquiring that evidence directly from WSIs. PathAgentBench provides a unified framework for measuring and improving evidence-seeking pathology models.

Songjun Tu, Chengdong Xu, Qichao Zhang et al.

Skill memories can improve agentic reinforcement learning by reusing past experience as textual guidance, but retrieved skills are not oracular: they may help in one state while misleading the same policy in another. This makes the common privileged-teacher assumption fragile, namely that a skill-conditioned prompt can be treated as a fixed teacher for the no-skill prompt. We introduce UCOB, a framework for learning to utilize and evolve agentic skills via credit-aware on-policy bidirectional self-distillation. UCOB treats skill-conditioned and no-skill prompts as two on-policy context views of the same model, compares their return-to-go within the same task and anchor state, and uses the higher-return view as the local teacher. This local credit signal internalizes useful skill-conditioned behavior, corrects misleading skill usage, and guides task/state skill memory updates, utility-aware retrieval, and reflection self-training. Experiments on agentic tasks, including ALFWorld, WebShop, and Search-QA, show that UCOB outperforms skill-free RL, skill-memory baselines, and self-distillation methods across model scales, with up to 23.5 and 18.0 point gains over SOTA baselines on ALFWorld and WebShop. Ablations and analyses further validate its core mechanisms and efficiency.

Zekai Zhang, Jiahao Li, Jie Zhang et al.

While text-to-image (T2I) models have achieved remarkable progress, they struggle with real-world requests that are often underspecified, implicit, or dependent on up-to-date knowledge. We identify this challenge as the Context Gap: the mismatch between the user context and the sufficient generation context for T2I models. To bridge this gap, we propose Qwen-Image-Agent, a unified agentic framework that integrates plan, reason, search, memory and feedback in a context-centric manner. Qwen-Image-Agent treats user input as partial context and progressively constructs the generation context through Context-Aware Planning and Context Grounding. Specifically, Context-Aware Planning identifies missing context and plans how it should be acquired and used, while Context Grounding gathers this context from reason, search, memory, and feedback. To evaluate agentic image generation, we further introduce Image Agent Bench (IA-Bench), a benchmark covering four core image agent capabilities: Plan, Reason, Search, and Memory. Experiments on IA-Bench, Mindbench and WISE-Verified show that Qwen-Image-Agent outperforms strong baselines and achieves state-of-the-art performance.

Hao Ding, Ling Liang, Ruitong Qiao et al.

Structured State Space Models (SSMs), such as Mamba, enable efficient long-sequence modeling with linear time complexity. Recent implementations realize this capability through Structured State Space Duality (SSD), which transforms recursive state evolution into matrix-form computations. However, SSD introduces substantial system-level overheads, including quadratic intermediate materialization, irregular data movement, and prefix-dependent execution, leading to excessive memory traffic and bandwidth demand on conventional architectures. Although prior accelerators mitigate these overheads through optimized dataflows or compute-in-memory techniques, they largely retain matrix-oriented SSD execution and cannot simultaneously avoid quadratic intermediate storage and efficiently map dependency-bound state propagation. This paper presents HEMERA, a heterogeneous memory-centric accelerator for efficient Mamba-2 inference. Rather than directly executing the matrix-form SSD computation, HEMERA reformulates it into an algebraically equivalent streaming-recursive dataflow that avoids quadratic intermediate storage while preserving the original computation. The resulting heterogeneous execution paradigm maps dense linear operations onto in-memory computing units and recursive state updates onto a dedicated streaming engine. Across Mamba-2 models ranging from 130M to 2.8B, HEMERA achieves average latency speedups of 1.4x-3.6x and energy-efficiency improvements of 12.2x-27.0x over the official optimized fused Mamba-2 kernel on NVIDIA A100. It further reduces the average SSD-related execution-time ratio across model scales to 14.12% during long-sequence inference, demonstrating its potential for efficient deployment under edge constraints.

Lingrui Li, Nan Pu, Dong Zhao et al.

Test-time adaptation (TTA) aims to mitigate distribution shifts by adapting models with unlabeled target data at inference time. While TTA with vision-language models (VLMs) has shown promising results in classification, extending it to medical image segmentation remains challenging. In this setting, the adaptation gains from optimizing on VLM-generated predictions are often outweighed by the degradation to the VLM's strong pretrained features caused by noisy, update-driven learning, resulting in limited and unstable improvements. We therefore propose Memory-Supported Synergistic Adaptation (MSSA), a novel training-free TTA framework for medical image segmentation. Without updating model parameters, MSSA dynamically selects reliable image-text predictions to construct an online memory, uses them as text-guided semantic priors, and couples them with cross-image structural alignment for robust adaptation. Specifically, MSSA consists of (i) a noise-aware memory construction module that filters and stabilizes cross-modal predictions, and (ii) a relevance-driven prototype alignment module that aligns the target sample with structurally consistent memory samples and their reliable predictions to improve adaptation. Extensive experiments on multiple medical segmentation benchmarks demonstrate that MSSA consistently improves VLM-based segmentation models and outperforms existing fine-tuning-based TTA methods by a clear margin, with gains of up to 12.2% DSC and 11.7% mIoU. Project page: https://lingrayy.github.io/MSSA/ .

Baogang Song, Zhongshu Zhao, Qianrong Zheng et al.

Biometric template protection (BTP) must secure stored templates while tolerating intra-class variations. Existing methods rely on protected-domain similarity matching, error correction, or predefined-template mappings, potentially retaining exploitable similarity structures, introducing helper-data risks, depending on artificial targets, or coupling protection to specific modalities. Storing only cryptographic hash digests eliminates directly comparable representations and conceals pre-hash templates, but hash-based exact-match verification requires genuine samples to generate identical intermediate templates before hashing. Identity classification is naturally suited to this requirement because it maps variable biometric samples to stable and discriminative identity-level outputs. Based on this insight, we propose Multiple Permuted-Label Classifier Encoding (MPLCE). Through classifier-specific label permutations, MPLCE assigns each identity different labels across multiple classifiers. The predicted labels are encoded and concatenated to form an intermediate template, preventing repeated encodings of a single identity label and enlarging the effective candidate space while preserving classification consistency. The template is randomized with an application-specific XOR string and cryptographically hashed, enabling exact-match verification without error correction codes or biometric-dependent helper data. Using modality-specific classifiers, MPLCE retains the same template generation and protection procedure across modalities. On four face and two iris datasets, MPLCE achieves competitive performance, including a GAR of 98.61\% at a FAR of 5.51\(\times\)10\textsuperscript{-5}\% on YTF and a GAR of 99.10\% at a FAR of 0.00\% on CASIA-Iris-Lamp. Security analyses and attack evaluations support its irreversibility, revocability, and unlinkability under the threat model.

Wenhao Li, Xueying Jiang, Quanhao Qian et al.

Real-world robot deployment rarely maintains the training-stage camera setup, where cameras often experience repositioning or remounting depending on actual scenarios. Existing view-robust Vision-Language-Action (VLA) policies tolerate such camera variations only when the camera extrinsics are explicitly provided, making them fragile and hard to use especially when view robustness is critical. We argue that the policy should not be told where the camera is, but rather figure it out by itself. To this end, we introduce Camera-Centric VLA (CamVLA), a new VLA model that decouples manipulation controls from camera geometry by predicting (i) a camera-centric end-effector action expressed in the local camera frame, and (ii) a 6-DoF hand-eye matrix relating cameras to the robot base. A deterministic geometric transformation composes the two predictions into a robot base-frame action. This disentangles how I should move in pose-independent camera-centric action generation from where I am looking from in camera-perspective geometric grounding. The resulting policy is calibration-free, depth-free, and single-view, requiring only a single monocular RGB image as the visual observation and task instruction at deployment. Evaluations in both simulation and real-world robot data show that CamVLA consistently improves success rates across diverse unseen viewpoints. Project page: https://alibaba-damo-academy.github.io/CamVLA/.

Haoyu Zhao, Xingyue Zhao, Siteng Huang et al.

Robotic manipulation in the open world requires not only recognizing what a scene looks like, but also anticipating how its 3D structure moves under interaction. We argue that synchronized RGB, depth, and optical flow, namely RGB-DF, provide a physically grounded representation that captures the underlying 4D dynamics of a scene. Compared to 2D pixel videos, this multi-modal synergy aligns visual appearance with geometric structure and temporal motion, creating a representation space significantly closer to the low-level end-effector actions demanded by robotic systems, thereby narrowing the gap between world prediction and policy learning. Building on this insight, we introduce RynnWorld-4D, a generative model that co-produces future RGB frames, depth maps, and optical flow from a single RGB-D image and a language instruction within one unified diffusion process. This 4D world model features a tri-branch architecture that integrates cross-modal attention with frame-wise 3D RoPE, ensuring that appearance, geometry, and motion evolve consistently. To supply training data at scale, we curate Rynn4DDataset 1.0, a massive dataset of over 254.4 million frames across egocentric human and robotic manipulation videos with high-quality pseudo-labels for depth and optical flow. We further propose RynnWorld-4D-Policy, an inverse dynamics head that consumes the internal 4D representations of RynnWorld-4D in a single forward pass, bypassing expensive multi-step denoising, to output robot actions in a closed-loop manner. Experiments show that RynnWorld-4D produces temporally and spatially coherent 4D predictions, and that RynnWorld-4D-Policy achieves state-of-the-art performance on real-world dexterous bimanual manipulation tasks, particularly excelling in tasks demanding spatial precision and temporal coordination.

Haoyu Zhao, Xingyue Zhao, Hangyu Li et al.

Scaling robot learning requires massive, diverse trajectory data, yet collection is currently bottlenecked by physical teleoperation, where every demonstration binds operator time to specific hardware and workspaces. We introduce digital teleoperation, a paradigm that decouples data collection from physical constraints by replacing the real robot with a generative world model. In this framework, an operator's hand-pose stream drives a robot-centric generative world model to synthesize high-fidelity egocentric videos from a single reference image. The recorded pose stream serves as an embodiment-agnostic action label transferable to any target robot via standard retargeting, yielding complete state-action trajectories for imitation learning independent of physical hardware. We instantiate this paradigm in RynnWorld-Teleop, a system that integrates depth-aware skeletal conditioning, progressive human-to-robot training on a video Diffusion Transformer, and streaming autoregressive distillation. This pipeline compresses the generative process into a single-pass inference, enabling 40+ FPS, real-time interactive generation on a single H100 GPU. Policies trained exclusively on RynnWorld-Teleop-generated data achieve effective zero-shot Sim2Real transfer across dexterous and diverse bimanual tasks. Moreover, augmenting real-world datasets with our digitally teleoperated data consistently improves success rates, demonstrating that RynnWorld-Teleop serves as a high-fidelity, scalable data engine for the next generation of robotic agents.

Bowen Jing, Mingxin Wang, Ruiyang Hao et al.

Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or drop while avoiding excessive deformation. However, existing manipulation benchmarks are predominantly success-oriented and rarely evaluate whether a policy remains physically safe throughout execution. We present SoftVTBench, a safety-aware visuo-tactile benchmark for physically constrained deformable object manipulation. Built in Isaac Sim with finite-element-simulated deformable objects, SoftVTBench provides multi-view RGB observations, RGB tactile sensing with marker motion, proprioception, and language instructions, and defines four matched task suites over object type (deformable vs. rigid) and variation axis (object vs. spatial). It separately reports Goal Success and Safety Success; the latter additionally requires no drop and peak deformation below a calibrated object-specific threshold, measured from policy-hidden privileged Finite Element Method (FEM) states. We implement pi0.5-based baselines under this protocol. Experiments show that success-only evaluation substantially overstates policy performance, as a large fraction of goal-completing rollouts still violate physical safety. Furthermore, incorporating tactile sensing improves Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation during execution, while maintaining comparable Goal Success. SoftVTBench provides a reproducible benchmark for studying visuo-tactile deformable manipulation under physical interaction constraints.

Dongdong Zhao, Jinrong Hu, Changtian Song et al.

Black-box membership inference attacks (MIAs) rely on target-model queries to infer whether candidate samples were used for training. However, membership signals are highly non-uniform across samples: some candidate samples support strong member/non-member separability, whereas many others provide little useful signal. Consequently, indiscriminate querying can incur substantial query cost and increase query-induced exposure, with limited marginal benefit for inference. This raises a key question: which candidate samples are worth querying for black-box MIAs? To address this question, we propose PSS-MIA, a pre-query sample selection framework which can be embedded with any existing MIA methods. PSS-MIA proceeds in two stages: it first ranks candidate samples and selects a subset expected to support stronger membership inference, then queries the selected samples and uses the returned outputs for an existing black-box MIA, thereby reducing query cost and query-induced exposure. In the first stage, we propose Loss-Gap Ranking (LGR), which ranks candidate samples by estimating the strength of their membership signal using loss gaps computed from reference models. Experiments on CIFAR-10, CIFAR-100, and CINIC-10 with five representative black-box MIA methods demonstrate that PSS-MIA with LGR consistently outperforms all other compared methods. Moreover, under a 0.1% FPR constraint, PSS-MIA can save at least 83.1%, 60.6%, and 80.4% of the query budget for the three datasets, respectively.

Dongfang Zhao

Graph-based approximate nearest neighbor (ANN) indexes on the GPU are originally built for static collections and must reconstruct the affected window to absorb any update, while the dynamic graph indexes that update incrementally run on the CPU and cannot exploit GPU parallelism. This paper presents a new ANN index, namely ETALE (Evolving Topology with Accelerated Lock-free Execution), which is among the first GPU-native graph ANN indexes to support streaming insertion and deletion without a global rebuild. The core of ETALE is a lock-free copy-on-write slab graph whose deletion state and adjacency share a single atomically published word, which yields a provable deletion-monotonicity invariant together with a bounded reclaim of GPU memory that is sublinear in accumulated deletions. We have implemented ETALE in CUDA and evaluate it on five diverse multimodal datasets against four state-of-the-art indexes: Tagore (GPU static, SIGMOD'26), DIGRA (CPU dynamic, SIGMOD'26), CAGRA (GPU static, ICDE'24), and HNSW (CPU dynamic, TPAMI'20). Under continuous churn, ETALE maintains the index in hundreds of milliseconds per round at recall above $0.95$, which is $4.8$--$8.8\times$ faster than CAGRA's per-window rebuild, $1.8$--$2.5\times$ faster than the more recent Tagore, and $3.3$--$147\times$ faster than the CPU dynamic indexes. In addition, the memory footprint of ETALE stays bounded where tombstone-only systems grow indefinitely.

Zhen Li, Xin Huang, Liang Wang et al.

LLM-based text embedders have substantially improved retrieval and semantic representation quality, but their deployment remains costly: large backbone models slow down embedding inference, while high-dimensional full-precision embeddings impose substantial storage and bandwidth overhead on large-scale indexes. In this paper, we present BITEMBED, an extreme low-bit framework for LLM-based text embedding that jointly targets encoding efficiency and vector storage. BITEMBED converts pretrained LLM backbones into BitNet-style embedding encoders with ternary weights, quantized activations, and lightweight normalization refinement. The converted model is adapted to representation learning through continual contrastive pre-training, followed by supervised contrastive fine-tuning with both similarity-distribution distillation and attention-relation distillation from a full-precision teacher. Beyond quantizing the backbone, BITEMBED further trains output embeddings to support multiple storage precisions meeting different storage needs in various scenarios. Experiments on MMTEB (eng, v2) with Qwen3-0.6B and Gemma3-270M show that BITEMBED is largely comparable to full precision teacher embedders. Moreover, BITEMBED flexibly obtains text embeddings of various precisions, achieving a trade-off between performance and storage cost.

Shiguo Lian, Kai Wang, Zhaoxiang Liu et al.

Large model inference optimization serves as a key foundation for supporting the scalable, low-cost, and highly stable operation of large model services. Centered on token-oriented inference optimization technology, this paper proposes for the first time a four-layer technical architecture consisting of Multi-model Fusion, Model Optimization, Compute-Model Fusion, and Compute-Network-Model Fusion. It systematically reviews the key technologies and current industry status across these four levels and analyzes the application value of related technologies in real-world business scenarios. This paper provides a practical technical path for reducing token production costs, improving token service efficiency, ensuring the stability of token supply, and driving the transition of large model services from being merely callable to being operable.

Changyi Lin, Ding Zhao

We present ART-Glove, an articulated tactile glove designed to capture contact-grounded dexterous demonstrations while preserving human dexterity. ART-Glove makes hand-side contact geometry explicit with 16 rigid functional surfaces covering the fingers, thumb, and palm. Twenty-two anatomically aligned joints connect these surfaces and allow them to follow human hand motion during dexterous manipulation. Encoder-based sensing tracks surface motion, while dense piezoresistive tactile sensing records contact over the same surfaces. The complete system captures synchronized 22-DoF joint measurements and 2048-taxel tactile measurements at 120 Hz. We evaluate ART-Glove across experiments on motion freedom, joint sensing, tactile sensing, and contact-rich interaction capture, demonstrating its ability to preserve human dexterity while recording contact-grounded information that can support downstream dexterous robot learning.