arXiv subjects

Zelin Peng

Publications and source records attributed to Zelin Peng.

At least 19 recordsLinked to original sources

arXiv2026

CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements

Despite recent advances in CUAs, their development remains constrained by the limited diversity, controllability, and reliability of interaction data. Moreover, conventional optimization pipelines lack iterative feedback, while existing recursive improvement approaches often rely on self-generated experiences, limiting ability expansion and yielding diminishing returns. To address these challenges, we introduce CUA-Master, a nested double-loop recursive improvement framework. At the system level, an outer world-data-model flywheel continuously incorporates open-world corpus and leverages evolving model weaknesses to guide subsequent data construction and optimization iterations. At the data level, an inner flywheel couples open-world seeding and autocatalytic propagation. The former mines authentic workflows as seed tasks from instructional videos, and employs visual-semantic dual-path trajectory reconstruction; The latter transforms these seeds into reusable primitives for controllable synthesis of diverse, complex, and verifiable tasks. To enable efficient and reliable trajectory collection, we further introduce a feedback-guided snapshot-rollback retry mechanism in various real environments. Extensive experiments demonstrate that CUA-Master-27B achieves competitive performance across GUI-only and GUI-CLI hybrid benchmarks, scoring 80.2 on OSWorld Verified, 36.9 on OSWorld 2.0, 43.9 on WeaveBench, and 25.0 on Agents' Last Exam. Our framework also substantially improves the quality and efficiency of data synthesis, and exhibits continuous gains across recursive iterations.

cs.AI↗

arXiv2026

Why Struggle with Continuous Latents? Interpretable Discrete Latent Reasoning via Rendered Compression

Large language models achieve high reasoning performance via explicit chain-of-thought and reinforcement learning, but require long output sequences and extended inference time. Latent reasoning reduces this cost by shifting computation into a latent space; however, continuous latent methods are hard to train, suffering from unstable and uninterpretable reasoning trajectories. We argue these issues stem from a misalignment between continuous-space reasoning and discrete symbolic supervision, as continuous states lack explicit anchors for step-by-step alignment. To resolve this, we propose \textbf{Discrete Latent Reasoning~(DLR)}, the first method that converts continuous latent states into explicit discrete tokens. Inspired by render-based compression, we render textual chains of thought into images, extract visual features, and construct a discrete latent vocabulary via clustering-based fine-tuning. Expanding the vocabulary and output head enables standard autoregressive modeling over both natural language and latent tokens, supporting pretraining alignment, SFT, and RL. Experiments on five reasoning benchmarks and two model series~(Qwen3-VL and LLaMA-3) confirm that \textbf{DLR} outperforms prior latent reasoning baselines with up to \textbf{20$\times$ compression}. Furthermore, the learned latent trajectories retain an interpretable semantic structure. Overall, discrete latent tokens provide a controllable and interpretable basis for efficient latent reasoning. Code is available at github.com/Miraclecsc/Discrete-Latent-Reasoning

cs.CL↗

arXiv2026

MIRA: A Multimodal Illustrated Deep Research Agent for Controllable Reports

Deep research agents are increasingly capable of searching the open web and synthesizing evidence, yet they remain poorly suited to controllable report generation in multimodal deep research. This task requires agents to handle interleaved textual and visual inputs, gather multimodal evidence through multi-step web search, and deliver comprehensive illustrated reports under user-specified requirements. Existing systems face two limitations. During reasoning, they often underuse images discovered along the search trajectory; during generation, they tend to produce fixed-form answers with limited control over report granularity, multimodal composition, and evidence grounding. We introduce MIRA, a Multimodal Illustrated Deep Research Agent for controllable reports. MIRA integrates textual and visual evidence throughout research and writing. At its core, MIRA-Harness translates user requirements into generation contracts and connects four stages: evidence collection, report submission, feedback-driven revision, and final delivery. Building on this harness, MIRA-Data constructs high-quality multi-hop multimodal research trajectories, and we post-train foundation models to operate within the workflow. Using Qwen3.6-35B-A3B, MIRA achieves state-of-the-art performance among open agents on 7 representative, established multimodal-search and deep-research-agent benchmarks, while producing user-aligned illustrated reports with traceable textual and visual evidence.

cs.CV↗

arXiv2026

HyperCLIP++: Fine-tuning CLIP forOpen-vocabulary Semantic Segmentation in Hyperbolic Space

CLIP, a foundational vision-language model, has emerged as a powerful tool for open-vocabulary semantic segmentation. While freezing CLIP's text encoder is known to preserve its generalization capability, recent studies show that fine-tuning both CLIP's text and image encoders jointly significantly enhances segmentation performance, especially for classes from open sets. In this work, we explain this phenomenon from the perspective of hierarchy alignment, since during fine-tuning, the hierarchical level of image embeddings shifts from image-level to pixel-level. We achieve this by leveraging hyperbolic space, which naturally encodes hierarchical structures. Our key observation is that, during fine-tuning, the hyperbolic radius of CLIP's text embeddings decreases, facilitating better alignment with the pixel-level granularity of visual data. Building on this, we propose HyperCLIP++, a novel and parameter-efficient adaptation strategy. HyperCLIP++ directly adjusts the hyperbolic radius of CLIP's embeddings via scaling transformations to achieve a hierarchy alignment to the target task, i.e., segmentation. To ensure this hierarchy alignment is effected consistently across both modalities and preserves their cross-modal alignment during training, HyperCLIP++ integrates a Dual Cross-Relation Communication (DCRC) module that synchronizes these adjustments between the vision and text pathways. Our experiments show that HyperCLIP++ achieves state-of-the-art performance across three benchmarks while fine-tuning only approximately 5% of CLIP's total parameters. More importantly, we observe that after adjustment, CLIP's text embeddings exhibit a relatively fixed hyperbolic radius across datasets, suggesting that the hierarchical level required for this segmentation task might be quantified using the hyperbolic radius.

cs.CV↗

arXiv2026

NEARL: Interacted Query Adaptation with Orthogonal Regularization for Medical Vision-Language Understanding

Computer-aided medical image analysis is crucial for disease diagnosis and treatment planning. While vision-language models (VLMs) such as CLIP exhibit strong generalization ability, their direct application to medical imaging remains hindered by a substantial domain gap. Existing methods for bridging this gap, including prompt learning and unidirectional modality interaction, typically introduce domain knowledge into only one modality. However, such approaches fail to fully exploit CLIP's inherent dual-modality structure and overlook the synergistic effect of bidirectional cross-modal interaction, resulting in persistent modality misalignment. In this paper, we propose NEARL (iNteracted quEry Adaptation with oRthogonaL Regularization), a novel parameter-efficient VLM framework for bidirectional cross-modal interaction. NEARL consists of two key components: (1) the Unified Synergy Embedding Transformer (USEformer), which dynamically generates compact cross-modal queries to facilitate interaction; and (2) the Orthogonal Cross-Attention Adapter (OCA), which decouples new knowledge into truly novel and incremental components through orthogonal regularization. This design reduces interference from incremental components, enabling more focused learning of novel information and improving modality interaction in VLMs. Notably, NEARL introduces only 1.46M learnable parameters. Extensive experiments on three medical imaging modalities demonstrate state-of-the-art performance (e.g., a 2.3% relative improvement on the pneumonia dataset), along with fast inference and low memory overhead, highlighting its effectiveness for real-world medical vision-language understanding.

cs.CV↗

arXiv2026

MMRad-22K: A Structured Multimodal Evidence Dataset for Chest X-ray Report Generation

Chest X-ray (CXR) reporting follows a region-based clinical workflow in which radiologists inspect anatomical regions and integrate localized findings into a final report. However, existing resources for CXR report generation provide these supervision signals in fragmented forms. We introduce MMRad-22K, a dataset that organizes regional textual observations, anatomical grounding coordinates, localized image evidence, and report targets into structured multimodal evidence units for CXR report generation. To motivate this formulation, we first compare different evidence formats for report generation and find that structured multimodal evidence is generally more useful than text-only or bounding box-based evidence. We then adapt a unified LVLM backbone using MMRad-22K and show that adaptation with multimodal evidence outperforms both textual-evidence adaptation and end-to-end adaptation on language and clinically oriented metrics. Under the same evaluation protocol, the adapted model also reaches a performance level comparable to several open-source LVLM references. Together, these results support MMRad-22K as a practical structured multimodal resource for training and evaluating CXR report generation aligned with clinical reading workflows.

cs.CV↗

arXiv2026

Seeking Consensus: Geometric-Semantic On-the-Fly Recalibration for Open-Vocabulary Remote Sensing Semantic Segmentation

Open-vocabulary semantic segmentation (OVSS) in remote sensing images is a promising task that employs textual descriptions for identifying undefined land cover categories. Despite notable advances, existing methods typically employ a static inference paradigm, overlooking the distinct distribution of each scene, resulting in semantic ambiguity in diverse land covers and incomplete foreground activation. Motivated by this, we propose Seeking Consensus, termed SeeCo, a plug-and-play framework to boost the performance of training-free OVSS models in remote sensing images, which recalibrates arbitrary OVSS models on-the-fly by seeking dual consensus: geometric consensus learning (GCL) through multi-view consistent observations and semantic consensus learning (SCL) via textual description adaptive calibration, which assists collaborative recalibration of visual and textual semantics. The two consensus are injected via an online consensus injector (OCI), effectively alleviating the under-activation and semantic bias. SeeCo requires no specific training process, yet recalibrates semantic-geometric alignment for each unique scene during inference. Extensive experiments on eight remote sensing OVSS benchmarks show consistent gains, proving its effectiveness and universality.

cs.CV↗

arXiv2026

Unlocking 3D Affordance Segmentation with 2D Semantic Knowledge

Affordance segmentation aims to decompose 3D objects into parts that serve distinct functional roles, enabling models to reason about object interactions rather than mere recognition. Existing methods, mostly following the paradigm of 3D semantic segmentation or prompt-based frameworks, struggle when geometric cues are weak or ambiguous, as sparse point clouds provide limited functional information. To overcome this limitation, we leverage the rich semantic knowledge embedded in large-scale 2D Vision Foundation Models (VFMs) to guide 3D representation learning through a cross-modal alignment mechanism. Specifically, we propose Cross-Modal Affinity Transfer (CMAT), a pretraining strategy that compels the 3D encoder to align with the semantic structures induced by lifted 2D features. CMAT is driven by a core affinity alignment objective, supported by two auxiliary losses, geometric reconstruction and feature diversity, which together encourage structured and discriminative feature learning. Built upon the CMAT-pretrained backbone, we employ a lightweight affordance segmentor that injects text or visual prompts into the learned 3D space through an efficient cross-attention interface, enabling dense and prompt-aware affordance prediction while preserving the semantic organization established during pretraining. Extensive experiments demonstrate consistent improvements over previous state-of-the-art methods in both accuracy and efficiency.

cs.CV↗

arXiv2026

AnimateScene: Camera-controllable Animation in Any Scene

Recent advances in 3D scene reconstruction and 4D human animation have broadened adoption, but integrating the two remains difficult. Key challenges include placing humans at plausible locations and scales without interpenetration, aligning lighting and style between humans and background, and handling dynamic camera trajectories. We introduce AnimateScene, a unified framework addressing these issues. First, an accurate placement module automatically determines realistic 3D positions and prevents collisions during motion. Second, we propose a training-free style alignment method that adapts the 4D human to match background illumination and appearance, achieving coherent composites. Finally, we design a joint post-reconstruction approach that incorporates camera trajectories for smooth, visually engaging motion videos. Experiments across diverse scenes and actions demonstrate that AnimateScene generates dynamic results with high geometric detail and strong spatiotemporal coherence. The project page is available at https://whynothaha.github.io/AnimateScene/AnimateScene.html.

cs.CV↗

arXiv2025

HyperET: Efficient Training in Hyperbolic Space for Multi-modal Large Language Models

Multi-modal large language models (MLLMs) have emerged as a transformative approach for aligning visual and textual understanding. They typically require extremely high computational resources (e.g., thousands of GPUs) for training to achieve cross-modal alignment at multi-granularity levels. We argue that a key source of this inefficiency lies in the vision encoders they widely equip with, e.g., CLIP and SAM, which lack the alignment with language at multi-granularity levels. To address this issue, in this paper, we leverage hyperbolic space, which inherently models hierarchical levels and thus provides a principled framework for bridging the granularity gap between visual and textual modalities at an arbitrary granularity level. Concretely, we propose an efficient training paradigm for MLLMs, dubbed as HyperET, which can optimize visual representations to align with their textual counterparts at an arbitrary granularity level through dynamic hyperbolic radius adjustment in hyperbolic space. HyperET employs learnable matrices with Möbius multiplication operations, implemented via three effective configurations: diagonal scaling matrices, block-diagonal matrices, and banded matrices, providing a flexible yet efficient parametrization strategy. Comprehensive experiments across multiple MLLM benchmarks demonstrate that HyperET consistently improves both existing pre-training and fine-tuning MLLMs clearly with less than 1\% additional parameters. Code is available at https://github.com/godlin-sjtu/HyperET

cs.CV↗

arXiv2025

OneCAT: Decoder-Only Auto-Regressive Model for Unified Understanding and Generation

We introduce OneCAT, a unified multimodal model that seamlessly integrates understanding, generation, and editing within a novel, pure decoder-only transformer architecture. Our framework uniquely eliminates the need for external components such as Vision Transformers (ViT) or vision tokenizer during inference, leading to significant efficiency gains, especially for high-resolution inputs. This is achieved through a modality-specific Mixture-of-Experts (MoE) structure trained with a single autoregressive (AR) objective, which also natively supports dynamic resolutions. Furthermore, we pioneer a multi-scale visual autoregressive mechanism within the Large Language Model (LLM) that drastically reduces decoding steps compared to diffusion-based methods while maintaining state-of-the-art performance. Our findings demonstrate the powerful potential of pure autoregressive modeling as a sufficient and elegant foundation for unified multimodal intelligence. As a result, OneCAT sets a new performance standard, outperforming existing open-source unified multimodal models across benchmarks for multimodal generation, editing, and understanding.

cs.CV↗

arXiv2025

MedSeg-R: Reasoning Segmentation in Medical Images with Multimodal Large Language Models

Medical image segmentation is crucial for clinical diagnosis, yet existing models are limited by their reliance on explicit human instructions and lack the active reasoning capabilities to understand complex clinical questions. While recent advancements in multimodal large language models (MLLMs) have improved medical question-answering (QA) tasks, most methods struggle to generate precise segmentation masks, limiting their application in automatic medical diagnosis. In this paper, we introduce medical image reasoning segmentation, a novel task that aims to generate segmentation masks based on complex and implicit medical instructions. To address this, we propose MedSeg-R, an end-to-end framework that leverages the reasoning abilities of MLLMs to interpret clinical questions while also capable of producing corresponding precise segmentation masks for medical images. It is built on two core components: 1) a global context understanding module that interprets images and comprehends complex medical instructions to generate multi-modal intermediate tokens, and 2) a pixel-level grounding module that decodes these tokens to produce precise segmentation masks and textual responses. Furthermore, we introduce MedSeg-QA, a large-scale dataset tailored for the medical image reasoning segmentation task. It includes over 10,000 image-mask pairs and multi-turn conversations, automatically annotated using large language models and refined through physician reviews. Experiments show MedSeg-R's superior performance across several benchmarks, achieving high segmentation accuracy and enabling interpretable textual analysis of medical images.

cs.CV↗

arXiv2025

Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding

Recent advancements in multimodal large language models (MLLMs) have significantly improved performance in visual question answering. However, they often suffer from hallucinations. In this work, hallucinations are categorized into two main types: initial hallucinations and snowball hallucinations. We argue that adequate contextual information can be extracted directly from the token interaction process. Inspired by causal inference in the decoding strategy, we propose to leverage causal masks to establish information propagation between multimodal tokens. The hypothesis is that insufficient interaction between those tokens may lead the model to rely on outlier tokens, overlooking dense and rich contextual cues. Therefore, we propose to intervene in the propagation process by tackling outlier tokens to enhance in-context inference. With this goal, we present FarSight, a versatile plug-and-play decoding strategy to reduce attention interference from outlier tokens merely by optimizing the causal mask. The heart of our method is effective token propagation. We design an attention register structure within the upper triangular matrix of the causal mask, dynamically allocating attention to capture attention diverted to outlier tokens. Moreover, a positional awareness encoding method with a diminishing masking rate is proposed, allowing the model to attend to further preceding tokens, especially for video sequence tasks. With extensive experiments, FarSight demonstrates significant hallucination-mitigating performance across different MLLMs on both image and video benchmarks, proving its effectiveness.

cs.CV↗

arXiv2025

ScalingNoise: Scaling Inference-Time Search for Generating Infinite Videos

Video diffusion models (VDMs) facilitate the generation of high-quality videos, with current research predominantly concentrated on scaling efforts during training through improvements in data quality, computational resources, and model complexity. However, inference-time scaling has received less attention, with most approaches restricting models to a single generation attempt. Recent studies have uncovered the existence of "golden noises" that can enhance video quality during generation. Building on this, we find that guiding the scaling inference-time search of VDMs to identify better noise candidates not only evaluates the quality of the frames generated in the current step but also preserves the high-level object features by referencing the anchor frame from previous multi-chunks, thereby delivering long-term value. Our analysis reveals that diffusion models inherently possess flexible adjustments of computation by varying denoising steps, and even a one-step denoising approach, when guided by a reward signal, yields significant long-term benefits. Based on the observation, we proposeScalingNoise, a plug-and-play inference-time search strategy that identifies golden initial noises for the diffusion sampling process to improve global content consistency and visual diversity. Specifically, we perform one-step denoising to convert initial noises into a clip and subsequently evaluate its long-term value, leveraging a reward model anchored by previously generated content. Moreover, to preserve diversity, we sample candidates from a tilted noise distribution that up-weights promising noises. In this way, ScalingNoise significantly reduces noise-induced errors, ensuring more coherent and spatiotemporally consistent video generation. Extensive experiments on benchmark datasets demonstrate that the proposed ScalingNoise effectively improves long video generation.

cs.LG↗

arXiv2025

ACMamba: Fast Unsupervised Anomaly Detection via An Asymmetrical Consensus State Space Model

Unsupervised anomaly detection in hyperspectral images (HSI), aiming to detect unknown targets from backgrounds, is challenging for earth surface monitoring. However, current studies are hindered by steep computational costs due to the high-dimensional property of HSI and dense sampling-based training paradigm, constraining their rapid deployment. Our key observation is that, during training, not all samples within the same homogeneous area are indispensable, whereas ingenious sampling can provide a powerful substitute for reducing costs. Motivated by this, we propose an Asymmetrical Consensus State Space Model (ACMamba) to significantly reduce computational costs without compromising accuracy. Specifically, we design an asymmetrical anomaly detection paradigm that utilizes region-level instances as an efficient alternative to dense pixel-level samples. In this paradigm, a low-cost Mamba-based module is introduced to discover global contextual attributes of regions that are essential for HSI reconstruction. Additionally, we develop a consensus learning strategy from the optimization perspective to simultaneously facilitate background reconstruction and anomaly compression, further alleviating the negative impact of anomaly reconstruction. Theoretical analysis and extensive experiments across eight benchmarks verify the superiority of ACMamba, demonstrating a faster speed and stronger performance over the state-of-the-art.

cs.CV↗

arXiv2025

PG-SAM: Prior-Guided SAM with Medical for Multi-organ Segmentation

Segment Anything Model (SAM) demonstrates powerful zero-shot capabilities; however, its accuracy and robustness significantly decrease when applied to medical image segmentation. Existing methods address this issue through modality fusion, integrating textual and image information to provide more detailed priors. In this study, we argue that the granularity of text and the domain gap affect the accuracy of the priors. Furthermore, the discrepancy between high-level abstract semantics and pixel-level boundary details in images can introduce noise into the fusion process. To address this, we propose Prior-Guided SAM (PG-SAM), which employs a fine-grained modality prior aligner to leverage specialized medical knowledge for better modality alignment. The core of our method lies in efficiently addressing the domain gap with fine-grained text from a medical LLM. Meanwhile, it also enhances the priors' quality after modality alignment, ensuring more accurate segmentation. In addition, our decoder enhances the model's expressive capabilities through multi-level feature fusion and iterative mask optimizer operations, supporting unprompted learning. We also propose a unified pipeline that effectively supplies high-quality semantic information to SAM. Extensive experiments on the Synapse dataset demonstrate that the proposed PG-SAM achieves state-of-the-art performance. Our code is released at https://github.com/logan-0623/PG-SAM.

cs.CV↗

arXiv2024

Parameter-efficient Fine-tuning in Hyperspherical Space for Open-vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation seeks to label each pixel in an image with arbitrary text descriptions. Vision-language foundation models, especially CLIP, have recently emerged as powerful tools for acquiring open-vocabulary capabilities. However, fine-tuning CLIP to equip it with pixel-level prediction ability often suffers three issues: 1) high computational cost, 2) misalignment between the two inherent modalities of CLIP, and 3) degraded generalization ability on unseen categories. To address these issues, we propose H-CLIP a symmetrical parameter-efficient fine-tuning (PEFT) strategy conducted in hyperspherical space for both of the two CLIP modalities. Specifically, the PEFT strategy is achieved by a series of efficient block-diagonal learnable transformation matrices and a dual cross-relation communication module among all learnable matrices. Since the PEFT strategy is conducted symmetrically to the two CLIP modalities, the misalignment between them is mitigated. Furthermore, we apply an additional constraint to PEFT on the CLIP text encoder according to the hyperspherical energy principle, i.e., minimizing hyperspherical energy during fine-tuning preserves the intrinsic structure of the original parameter space, to prevent the destruction of the generalization ability offered by the CLIP text encoder. Extensive evaluations across various benchmarks show that H-CLIP achieves new SOTA open-vocabulary semantic segmentation results while only requiring updating approximately 4% of the total parameters of CLIP.

cs.CV↗

arXiv2024

OphCLIP: Hierarchical Retrieval-Augmented Learning for Ophthalmic Surgical Video-Language Pretraining

Surgical practice involves complex visual interpretation, procedural skills, and advanced medical knowledge, making surgical vision-language pretraining (VLP) particularly challenging due to this complexity and the limited availability of annotated data. To address the gap, we propose OphCLIP, a hierarchical retrieval-augmented vision-language pretraining framework specifically designed for ophthalmic surgical workflow understanding. OphCLIP leverages the OphVL dataset we constructed, a large-scale and comprehensive collection of over 375K hierarchically structured video-text pairs with tens of thousands of different combinations of attributes (surgeries, phases/operations/actions, instruments, medications, as well as more advanced aspects like the causes of eye diseases, surgical objectives, and postoperative recovery recommendations, etc). These hierarchical video-text correspondences enable OphCLIP to learn both fine-grained and long-term visual representations by aligning short video clips with detailed narrative descriptions and full videos with structured titles, capturing intricate surgical details and high-level procedural insights, respectively. Our OphCLIP also designs a retrieval-augmented pretraining framework to leverage the underexplored large-scale silent surgical procedure videos, automatically retrieving semantically relevant content to enhance the representation learning of narrative videos. Evaluation across 11 datasets for phase recognition and multi-instrument identification shows OphCLIP's robust generalization and superior performance.

cs.CV↗