Yifan Li

Yifan Li, Kun Zhou, Wayne Xin Zhao et al.

Non-autoregressive (NAR) text generation has attracted much attention in the field of natural language processing, which greatly reduces the inference latency but has to sacrifice the generation accuracy. Recently, diffusion models, a class of latent variable generative models, have been introduced into NAR text generation, showing an improved text generation quality. In this survey, we review the recent progress in diffusion models for NAR text generation. As the background, we first present the general definition of diffusion models and the text diffusion models, and then discuss their merits for NAR generation. As the core content, we further introduce two mainstream diffusion models in existing work of text diffusion, and review the key designs of the diffusion process. Moreover, we discuss the utilization of pre-trained language models (PLMs) for text diffusion models and introduce optimization techniques for text data. Finally, we discuss several promising directions and conclude this paper. Our survey aims to provide researchers with a systematic reference of related research on text diffusion models for NAR generation. We present our collection of text diffusion models at https://github.com/RUCAIBox/Awesome-Text-Diffusion-Models.

Wayne Xin Zhao, Kun Zhou, Junyi Li et al.

Language is essentially a complex, intricate system of human expressions governed by grammatical rules. It poses a significant challenge to develop capable AI algorithms for comprehending and grasping a language. As a major approach, language modeling has been widely studied for language understanding and generation in the past two decades, evolving from statistical language models to neural language models. Recently, pre-trained language models (PLMs) have been proposed by pre-training Transformer models over large-scale corpora, showing strong capabilities in solving various NLP tasks. Since researchers have found that model scaling can lead to performance improvement, they further study the scaling effect by increasing the model size to an even larger size. Interestingly, when the parameter scale exceeds a certain level, these enlarged language models not only achieve a significant performance improvement but also show some special abilities that are not present in small-scale language models. To discriminate the difference in parameter scale, the research community has coined the term large language models (LLM) for the PLMs of significant size. Recently, the research on LLMs has been largely advanced by both academia and industry, and a remarkable progress is the launch of ChatGPT, which has attracted widespread attention from society. The technical evolution of LLMs has been making an important impact on the entire AI community, which would revolutionize the way how we develop and use AI algorithms. In this survey, we review the recent advances of LLMs by introducing the background, key findings, and mainstream techniques. In particular, we focus on four major aspects of LLMs, namely pre-training, adaptation tuning, utilization, and capacity evaluation. Besides, we also summarize the available resources for developing LLMs and discuss the remaining issues for future directions.

Wayne Xin Zhao, Kun Zhou, Junyi Li et al.

Language is essentially a complex, intricate system of human expressions governed by grammatical rules. It poses a significant challenge to develop capable AI algorithms for comprehending and grasping a language. As a major approach, language modeling has been widely studied for language understanding and generation in the past two decades, evolving from statistical language models to neural language models. Recently, pre-trained language models (PLMs) have been proposed by pre-training Transformer models over large-scale corpora, showing strong capabilities in solving various NLP tasks. Since researchers have found that model scaling can lead to performance improvement, they further study the scaling effect by increasing the model size to an even larger size. Interestingly, when the parameter scale exceeds a certain level, these enlarged language models not only achieve a significant performance improvement but also show some special abilities that are not present in small-scale language models. To discriminate the difference in parameter scale, the research community has coined the term large language models (LLM) for the PLMs of significant size. Recently, the research on LLMs has been largely advanced by both academia and industry, and a remarkable progress is the launch of ChatGPT, which has attracted widespread attention from society. The technical evolution of LLMs has been making an important impact on the entire AI community, which would revolutionize the way how we develop and use AI algorithms. In this survey, we review the recent advances of LLMs by introducing the background, key findings, and mainstream techniques. In particular, we focus on four major aspects of LLMs, namely pre-training, adaptation tuning, utilization, and capacity evaluation. Besides, we also summarize the available resources for developing LLMs and discuss the remaining issues for future directions.

Chensheng Dai, Shengjun Zhang, Yifan Li et al.

Video generation models based on Diffusion Transformers (DiTs) have achieved remarkable performance in video synthesis, yet they suffer from high inference latency and computational costs due to the quadratic complexity of 3D attention. Existing acceleration methods primarily reduce computational complexity within each individual denoising steps through techniques such as sparse attention and KV-caching. However, they rigidly adhere to the inherent constraint of the standard diffusion pipeline: every frame in the target video sequence must be subjected to a complete, dense denoising process across all diffusion timesteps. We observe that due to the corresponding contents and motions among adjacent frames, when keyframes with critical semantic transitions are anchored, the intermediate states of others often follow more predictable trajectories, which indicates that such uniform, dense denoising process is inherently redundant for natural video data. To this end, we introduce \textbf{RhymeFlow}, a training-free framework that decouples the denoising trajectories of different frames. Specifically, we first identify a sparse set of pivotal key frames that dominate the latent semantic evolution. Then, only these keyframes undergo dense, step-by-step denoising to ensure structural integrity, while non-keyframes progressively skip denoising steps to minimize computational cost. Since skipped intermediate states of non-keyframes break the temporal coherence in keyframe denoising steps, leading to visual degradation, we further introduce a latent trajectory projection module, which enables keyframes to interact with a complete and temporally consistent sequence representation. Extensive experiments on current DiT-based video generation models demonstrate our method outperforms existing baselines with higher inference speed and better visual quality.

Zhaoshan Liu, Qiujie Lv, Yifan Li et al.

Data augmentation (DA) has been widely leveraged in computer vision to alleviate the data shortage, whereas the DA in medical image analysis (MIA) faces multiple challenges. The prevalent DA approaches in MIA encompass conventional DA, synthetic DA, and automatic DA. However, utilizing these approaches poses various challenges such as experience-driven design and intensive computation cost. Here, we propose an efficient and effective automatic DA method termed MedAugment. We propose a pixel augmentation space and spatial augmentation space and exclude the operations that can break medical details and features, such as severe color distortions or structural alterations that can compromise image diagnostic value. Besides, we propose a novel sampling strategy by sampling a limited number of operations from the two spaces. Moreover, we present a hyperparameter mapping relationship to produce a rational augmentation level and make the MedAugment fully controllable using a single hyperparameter. These configurations settle the differences between natural and medical images, such as high sensitivity to certain attributes such as brightness and posterize. Extensive experimental results on four classification and four segmentation datasets demonstrate the superiority of MedAugment. Compared with existing approaches, the proposed MedAugment serves as a more suitable yet general processing pipeline for medical images without producing color distortions or structural alterations and involving negligible computational overhead. We emphasize that our method can serve as a plugin for arbitrary projects without any extra training stage, thereby holding the potential to make a valuable contribution to the medical field, particularly for medical experts without a solid foundation in deep learning. Code is available at https://github.com/NUS-Tim/MedAugment.

Pengfei Zhou, Shengcong Chen, Di Chen et al.

Robotic manipulation requires models that generate executable actions while anticipating and evaluating their future consequences before physical execution. We present $τ_0$-World Model ($τ_0$-WM), a unified video-action world model that integrates policy learning, video prediction, and action evaluation within a single future-predictive framework. Built on a shared video diffusion backbone, $τ_0$-WM provides two complementary interfaces. First, a video action model jointly predicts future visual latents and continuous action chunks from multi-view observations, language instructions, and robot state. Second, an action-conditioned video simulator rolls out candidate action chunks into multi-view futures and predicts dense task-progress scores. The model is trained on approximately $27{,}300$ hours of real-robot teleoperation, UMI-style interaction, egocentric human videos, and rollout or failure trajectories using modality-specific supervision masks. At inference time, $τ_0$-WM uses test-time computation to sample action candidates, rank them with re-denoising consistency, and invoke simulator-based rectification for low-quality candidates. On challenging long-horizon and fine-grained robotic manipulation tasks, $τ_0$-WM shows superior performance over other relevant baselines.

Yifan Li, Yikai Wang, Yanwei Fu et al.

Visual-Language Alignment (VLA) has gained a lot of attention since CLIP's groundbreaking work. Although CLIP performs well, the typical direct latent feature alignment lacks clarity in its representation and similarity scores. On the other hand, lexical representation, a vector whose element represents the similarity between the sample and a word from the vocabulary, is a natural sparse representation and interpretable, providing exact matches for individual words. However, lexical representations are difficult to learn due to no ground-truth supervision and false-discovery issues, and thus requires complex design to train effectively. In this paper, we introduce LexVLA, a more interpretable VLA framework by learning a unified lexical representation for both modalities without complex design. We use DINOv2 as our visual model for its local-inclined features and Llama 2, a generative language model, to leverage its in-context lexical prediction ability. To avoid the false discovery, we propose an overuse penalty to refrain the lexical representation from falsely frequently activating meaningless words. We demonstrate that these two pre-trained uni-modal models can be well-aligned by fine-tuning on the modest multi-modal dataset and avoid intricate training configurations. On cross-modal retrieval benchmarks, LexVLA, trained on the CC-12M multi-modal dataset, outperforms baselines fine-tuned on larger datasets (e.g., YFCC15M) and those trained from scratch on even bigger datasets (e.g., 1.1B data, including CC-12M). We conduct extensive experiments to analyze LexVLA. Codes are available at https://github.com/Clementine24/LexVLA.

Shengjun Zhang, Zhang Zhang, Simin Huang et al.

Recent advancements in video-based world models have demonstrated an unprecedented ability to synthesize high-fidelity visual sequences. However, a fundamental gap persists between visually plausible video generation and the functional requirements of a world model, particularly in maintaining a stable and reasonable internal state over extended temporal horizons. While existing benchmarks primarily emphasize visual quality, motion coherence, and text-video alignment, they largely overlook memory, the core capability of a world model to preserve consistency across long-term horizons and complex interactions. To address this gap, we present \textbf{MBench}, a comprehensive benchmark dedicated to quantifying and evaluating the memory capability of video world models. We systematically decompose the memory capability of video world models into three hierarchical and complementary core dimensions: entity consistency, environment consistency, and causal consistency, which are further refined into 12 quantifiable sub-dimensions for comprehensive characterization of long-term memory. Our benchmark is built upon rigorously curated real-captured long videos, and evaluated by rule-based quantitative matrices and VLM to enable objective and comprehensive consistency assessment. Extensive evaluations of mainstream state-of-the-art video world models reveal critical systemic limitations of existing methods in long-term state retention, providing a standardized benchmark and clear research direction to advance the field.

Tianyu Wang, Yifan Li, Haitao Lin et al.

Enabling robots to understand language instructions and react accordingly to visual perception has been a long-standing goal in the robotics research community. Achieving this goal requires cutting-edge advances in natural language processing, computer vision, and robotics engineering. Thus, this paper mainly investigates the potential of integrating the most recent Large Language Models (LLMs) and existing visual grounding and robotic grasping system to enhance the effectiveness of the human-robot interaction. We introduce the WALL-E (Embodied Robotic WAiter load lifting with Large Language model) as an example of this integration. The system utilizes the LLM of ChatGPT to summarize the preference object of the users as a target instruction via the multi-round interactive dialogue. The target instruction is then forwarded to a visual grounding system for object pose and size estimation, following which the robot grasps the object accordingly. We deploy this LLM-empowered system on the physical robot to provide a more user-friendly interface for the instruction-guided grasping task. The further experimental results on various real-world scenarios demonstrated the feasibility and efficacy of our proposed framework. See the project website at: https://star-uu-wang.github.io/WALL-E/

Yifan Li, Haomiao Sun, Zhaori Liu et al.

Affective behaviour analysis has aroused researchers' attention due to its broad applications. However, it is labor exhaustive to obtain accurate annotations for massive face images. Thus, we propose to utilize the prior facial information via Masked Auto-Encoder (MAE) pretrained on unlabeled face images. Furthermore, we combine MAE pretrained Vision Transformer (ViT) and AffectNet pretrained CNN to perform multi-task emotion recognition. We notice that expression and action unit (AU) scores are pure and intact features for valence-arousal (VA) regression. As a result, we utilize AffectNet pretrained CNN to extract expression scores concatenating with expression and AU scores from ViT to obtain the final VA features. Moreover, we also propose a co-training framework with two parallel MAE pretrained ViT for expression recognition tasks. In order to make the two views independent, we random mask most patches during the training process. Then, JS divergence is performed to make the predictions of the two views as consistent as possible. The results on ABAW4 show that our methods are effective.

Zhaoshan Liu, Qiujie Lv, Ziduo Yang et al.

The transformer is primarily used in the field of natural language processing. Recently, it has been adopted and shows promise in the computer vision (CV) field. Medical image analysis (MIA), as a critical branch of CV, also greatly benefits from this state-of-the-art technique. In this review, we first recap the core component of the transformer, the attention mechanism, and the detailed structures of the transformer. After that, we depict the recent progress of the transformer in the field of MIA. We organize the applications in a sequence of different tasks, including classification, segmentation, captioning, registration, detection, enhancement, localization, and synthesis. The mainstream classification and segmentation tasks are further divided into eleven medical image modalities. A large number of experiments studied in this review illustrate that the transformer-based method outperforms existing methods through comparisons with multiple evaluation metrics. Finally, we discuss the open challenges and future opportunities in this field. This task-modality review with the latest contents, detailed information, and comprehensive comparison may greatly benefit the broad MIA community.

Angen Ye, Boyuan Wang, Chaojun Ni et al.

World-Action Models (WAM) initialized from pre-trained video generation backbones have demonstrated remarkable potential for robot policy learning. However, existing approaches face two critical bottlenecks that hinder performance and deployment. First, jointly reasoning over future visual dynamics and corresponding actions incurs substantial inference overhead. Second, joint modeling often entangles visual and motion representations, making motion prediction accuracy heavily dependent on the quality of future video forecasts. To address these issues, we introduce GigaWorld-Policy, an action-centered WAM that learns 2D pixel-action dynamics while enabling efficient action decoding, with optional video generation. Specifically, we formulate policy training into two coupled components: the model predicts future action sequences conditioned on the current observation, and simultaneously generates future videos conditioned on the predicted actions and the same observation. The policy is supervised by both action prediction and video generation, providing richer learning signals and encouraging physically plausible actions through visual-dynamics constraints. With a causal design that prevents future-video tokens from influencing action tokens, explicit future-video generation is optional at inference time, allowing faster action prediction during deployment. To support this paradigm, we curate a diverse, large-scale robot dataset to pre-train an action-centered video generation model, which is then adapted as the backbone for robot policy learning. Experimental results on real-world robotic platforms show that GigaWorld-Policy runs 9x faster than the leading WAM baseline, Motus, while improving task success rates by 7%. Moreover, compared with pi-0.5, GigaWorld-Policy improves performance by 95% on RoboTwin 2.0.

Yifan Li, Yukai Gu, Yingqian Min et al.

Recent breakthroughs in video generation have demonstrated an emerging capability termed Chain-of-Frames (CoF) reasoning, where models resolve complex tasks through the generation of continuous frames. While these models show promise for Generative Video Reasoning (GVR), existing evaluation frameworks often rely on single-frame assessments, which can lead to outcome-hacking, where a model reaches a correct conclusion through an erroneous process. To address this, we propose a process-aware evaluation paradigm. We introduce VIPER, a comprehensive benchmark spanning 16 tasks across temporal, structural, symbolic, spatial, physics, and planning reasoning. Furthermore, we propose Process-outcome Consistency (POC@r), a new metric that utilizes VLM-as-Judge with a hierarchical rubric to evaluate both the validity of the intermediate steps and the final result. Our experiments reveal that state-of-the-art video models achieve POC@1.0 only about 20% and exhibit a significant outcome-hacking. We further explore the impact of test-time scaling and sampling robustness, highlighting a substantial gap between current video generation and true generalized visual reasoning. Our benchmark are released at https://github.com/RUCAIBox/VIPER.

Yuxin Wen, Qing Shuai, Di Kang et al.

We present HY-Motion 1.0, a series of state-of-the-art, large-scale, motion generation models capable of generating 3D human motions from textual descriptions. HY-Motion 1.0 represents the first successful attempt to scale up Diffusion Transformer (DiT)-based flow matching models to the billion-parameter scale within the motion generation domain, delivering instruction-following capabilities that significantly outperform current open-source benchmarks. Uniquely, we introduce a comprehensive, full-stage training paradigm -- including large-scale pretraining on over 3,000 hours of motion data, high-quality fine-tuning on 400 hours of curated data, and reinforcement learning from both human feedback and reward models -- to ensure precise alignment with the text instruction and high motion quality. This framework is supported by our meticulous data processing pipeline, which performs rigorous motion cleaning and captioning. Consequently, our model achieves the most extensive coverage, spanning over 200 motion categories across 6 major classes. We release HY-Motion 1.0 to the open-source community to foster future research and accelerate the transition of 3D human motion generation models towards commercial maturity.

Ji Zhang, Xijun Li, Xiyao Zhou et al.

Cache plays an important role to maintain high and stable performance (i.e. high throughput, low tail latency and throughput jitter) in storage systems. Existing rule-based cache management methods, coupled with engineers' manual configurations, cannot meet ever-growing requirements of both time-varying workloads and complex storage systems, leading to frequent cache overloading. In this paper, we for the first time propose a light-weight learning-based cache bandwidth control technique, called \LQoCo which can adaptively control the cache bandwidth so as to effectively prevent cache overloading in storage systems. Extensive experiments with various workloads on real systems show that LQoCo, with its strong adaptability and fast learning ability, can adapt to various workloads to effectively control cache bandwidth, thereby significantly improving the storage performance (e.g. increasing the throughput by 10\%-20\% and reducing the throughput jitter and tail latency by 2X-6X and 1.5X-4X, respectively, compared with two representative rule-based methods).

Yifan Li, Pei Cheng, Bin Fu et al.

Video chroma-lux editing, which aims to modify illumination and color while preserving structural and temporal fidelity, remains a significant challenge. Existing methods typically rely on expensive supervised training with synthetic paired data. This paper proposes VibeFlow, a novel self-supervised framework that unleashes the intrinsic physical understanding of pre-trained video generation models. Instead of learning color and light transitions from scratch, we introduce a disentangled data perturbation pipeline that enforces the model to adaptively recombine structure from source videos and color-illumination cues from reference images, enabling robust disentanglement in a self-supervised manner. Furthermore, to rectify discretization errors inherent in flow-based models, we introduce Residual Velocity Fields alongside a Structural Distortion Consistency Regularization, ensuring rigorous structural preservation and temporal coherence. Our framework eliminates the need for costly training resources and generalizes in a zero-shot manner to diverse applications, including video relighting, recoloring, low-light enhancement, day-night translation, and object-specific color editing. Extensive experiments demonstrate that VibeFlow achieves impressive visual quality with significantly reduced computational overhead. Our project is publicly available at https://lyf1212.github.io/VibeFlow-webpage.

Yifan Du, Zikang Liu, Yifan Li et al.

Recently, slow-thinking reasoning systems, built upon large language models (LLMs), have garnered widespread attention by scaling the thinking time during inference. There is also growing interest in adapting this capability to multimodal large language models (MLLMs). Given that MLLMs handle more complex data semantics across different modalities, it is intuitively more challenging to implement multimodal slow-thinking systems. To address this issue, in this paper, we explore a straightforward approach by fine-tuning a capable MLLM with a small amount of textual long-form thought data, resulting in a multimodal slow-thinking system, Virgo (Visual reasoning with long thought). We find that these long-form reasoning processes, expressed in natural language, can be effectively transferred to MLLMs. Moreover, it seems that such textual reasoning data can be even more effective than visual reasoning data in eliciting the slow-thinking capacities of MLLMs. While this work is preliminary, it demonstrates that slow-thinking capacities are fundamentally associated with the language model component, which can be transferred across modalities or domains. This finding can be leveraged to guide the development of more powerful slow-thinking reasoning systems. We release our resources at https://github.com/RUCAIBox/Virgo.

Jinhu Qi, Yifan Li, Minghao Zhao et al.

As agentic AI systems move beyond static question answering into open-ended, tool-augmented, and multi-step real-world workflows, their increased authority poses greater risks of system misuse and operational failures. However, current evaluation practices remain fragmented, measuring isolated capabilities such as coding, hallucination, jailbreak resistance, or tool use in narrowly defined settings. We argue that the central limitation is not merely insufficient coverage of evaluation dimensions, but the lack of a principled notion of representativeness: an agent's trustworthiness should be assessed over a representative socio-technical scenario distribution rather than a collection of disconnected benchmark instances. To this end, we propose the Holographic Agent Assessment Framework (HAAF), a systematic evaluation paradigm that characterizes agent trustworthiness over a scenario manifold spanning task types, tool interfaces, interaction dynamics, social contexts, and risk levels. The framework integrates four complementary components: (i) static cognitive and policy analysis, (ii) interactive sandbox simulation, (iii) social-ethical alignment assessment, and (iv) a distribution-aware representative sampling engine that jointly optimizes coverage and risk sensitivity -- particularly for rare but high-consequence tail risks that conventional benchmarks systematically overlook. These components are connected through an iterative Trustworthy Optimization Factory. Through cycles of red-team probing and blue-team hardening, this paradigm progressively narrows the vulnerabilities to meet deployment standards, shifting agent evaluation from benchmark islands toward representative, real-world trustworthiness. Code and data for the illustrative instantiation are available at https://github.com/TonyQJH/haaf-pilot.

Yifan Li, Zhixin Lai, Wentao Bao et al.

Visual-language models (VLM) have emerged as a powerful tool for learning a unified embedding space for vision and language. Inspired by large language models, which have demonstrated strong reasoning and multi-task capabilities, visual large language models (VLLMs) are gaining increasing attention for building general-purpose VLMs. Despite the significant progress made in VLLMs, the related literature remains limited, particularly from a comprehensive application perspective, encompassing generalized and specialized applications across vision (image, video, depth), action, and language modalities. In this survey, we focus on the diverse applications of VLLMs, examining their using scenarios, identifying ethics consideration and challenges, and discussing future directions for their development. By synthesizing these contents, we aim to provide a comprehensive guide that will pave the way for future innovations and broader applications of VLLMs. The paper list repository is available: https://github.com/JackYFL/awesome-VLLMs.

GigaBrain Team, Boyuan Wang, Bohan Li et al.

Vision-language-action (VLA) models that directly predict multi-step action chunks from current observations face inherent limitations due to constrained scene understanding and weak future anticipation capabilities. In contrast, video world models pre-trained on web-scale video corpora exhibit robust spatiotemporal reasoning and accurate future prediction, making them a natural foundation for enhancing VLA learning. Therefore, we propose \textit{GigaBrain-0.5M*}, a VLA model trained via world model-based reinforcement learning. Built upon \textit{GigaBrain-0.5}, which is pre-trained on over 10,000 hours of robotic manipulation data, whose intermediate version currently ranks first on the international RoboChallenge benchmark. \textit{GigaBrain-0.5M*} further integrates world model-based reinforcement learning via \textit{RAMP} (Reinforcement leArning via world Model-conditioned Policy) to enable robust cross-task adaptation. Empirical results demonstrate that \textit{RAMP} achieves substantial performance gains over the RECAP baseline, yielding improvements of approximately 30\% on challenging tasks including \texttt{Laundry Folding}, \texttt{Box Packing}, and \texttt{Espresso Preparation}. Critically, \textit{GigaBrain-0.5M$^*$} exhibits reliable long-horizon execution, consistently accomplishing complex manipulation tasks without failure as validated by real-world deployment videos on our \href{https://gigabrain05m.github.io}{project page}.

Dawei Li, Zhen Tan, Peijia Qian et al.

While multi-agent systems have been shown to significantly enhance the performance of Large Language Models (LLMs) across various tasks and applications, the dense interaction between scaling agents potentially hampers their efficiency and diversity. To address these challenges, we draw inspiration from the sparse mixture-of-agents (SMoE) and propose a sparse mixture-of-agents (SMoA) framework to improve the efficiency and diversity of multi-agent LLMs. Unlike completely connected structures, SMoA introduces novel Response Selection and Early Stopping mechanisms to sparsify information flows among individual LLM agents, striking a balance between performance and efficiency. Additionally, inspired by the expert diversity principle in SMoE frameworks for workload balance between experts, we assign distinct role descriptions to each LLM agent, fostering diverse and divergent thinking. Extensive experiments on reasoning, alignment, and fairness benchmarks demonstrate that SMoA achieves performance comparable to traditional mixture-of-agents approaches but with significantly lower computational costs. Further analysis reveals that SMoA is more stable, has a greater capacity to scale, and offers considerable potential through hyper-parameter optimization. Code and data will be available at: https://github.com/David-Li0406/SMoA.

Yifan Li, Xinyu Zhou, Yunhao Ge et al.

Robotic manipulation is often specified through language instructions or task identifiers, yet cluttered environments with similar objects are better handled by spatially indicating what to move and where to place it. Addressing the vision-centric challenge of object and goal specification, we present, to the best of our knowledge, the first formalization of Spatially Prompted Visual Trajectory Prediction (SP-VTP). This novel setting utilizes initial spatial prompts (like bounding boxes or points) to define task objectives, tasking the model with forecasting future end-effector trajectories from egocentric streams. To study this problem, we collect and annotate EgoSPT, a dataset of egocentric spatially prompted manipulation trajectories with first-frame object and target grounding annotations and recovered 3D end-effector motion. SP-VTP is challenging because the task specification is static, while the scene configuration evolves over time. To solve this problem, we propose SPOT(Spatially Prompted Object-Target Policy), which combines a task encoder for first-frame visual and coordinate spatial prompts, an observation encoder for current visual and history context, and a trajectory generator for future end-effector motion. Experiments under strict scene-level splits show that SPOT improves cross-scene trajectory prediction over non-prompted or single-source prompted baselines. Together, EgoSPT and SPOT establish a new spatial prompting problem SP-VTP, as a simple and scalable task condition for egocentric manipulation.

Yifan Li, Seunghyun Yoon, Viet Dac Lai et al.

Vision Transformers (ViTs) based vision foundation models (VFMs) have achieved remarkable performance across diverse vision tasks, but suffer from quadratic complexity that limits scalability to long sequences. Existing linear attention approaches for ViTs are typically trained from scratch, requiring substantial computational resources, while linearization-based methods developed for large language model decoders do not transfer well to ViTs. To address these challenges, we propose ViT-AdaLA, a novel framework for effectively adapting and transferring prior knowledge from VFMs to linear attention ViTs. ViT-AdaLA consists of three stages: attention alignment, feature alignment, and supervised fine-tuning. In the attention alignment stage, we align vanilla linear attention with the original softmax-based attention in each block to approximate the behavior of softmax attention. However, residual approximation errors inevitably accumulate across layers. We mitigate this by fine-tuning the linearized ViT to align its final-layer features with a frozen softmax VFM teacher. Finally, the adapted prior knowledge is transferred to downstream tasks through supervised fine-tuning. Extensive experiments on classification and segmentation tasks demonstrate the effectiveness and generality of ViT-AdaLA over various state-of-the-art linear attention counterpart.

Zhongyi Cai, Bryce Gernon, Wentao Bao et al.

The proliferation of face forgeries has increasingly undermined confidence in the authenticity of online content. Given the rapid development of face forgery generation algorithms, new fake categories are likely to keep appearing, posing a major challenge to existing face forgery detection methods. Despite recent advances in face forgery detection, existing methods are typically limited to binary Real-vs-Fake classification or the identification of known fake categories, and are incapable of detecting the emergence of novel types of forgeries. In this work, we study the Open Set Face Forgery Detection (OSFFD) problem, which demands that the detection model recognize novel fake categories. We reformulate the OSFFD problem and address it through uncertainty estimation, enhancing its applicability to real-world scenarios. Specifically, we propose the Dual-Level Evidential face forgery Detection (DLED) approach, which collects and fuses category-specific evidence on the spatial and frequency levels to estimate prediction uncertainty. Extensive evaluations conducted across diverse experimental settings demonstrate that the proposed DLED method achieves state-of-the-art performance, outperforming various baseline models by an average of 20% in detecting forgeries from novel fake categories. Moreover, on the traditional Real-versus-Fake face forgery detection task, our DLED method concurrently exhibits competitive performance.

Yingqian Min, Kun Zhou, Yifan Li et al.

Recent advancements in reinforcement learning with verifiable rewards (RLVR) have significantly improved the complex reasoning ability of vision-language models (VLMs). However, its outcome-level supervision is too coarse to diagnose and correct errors within the reasoning chain. To this end, we propose Perceval, a process reward model (PRM) that enables token-level error grounding, which can extract image-related claims from the response and compare them one by one with the visual evidence in the image, ultimately returning claims that contain perceptual errors. Perceval is trained with perception-intensive supervised training data. We then integrate Perceval into the RL training process to train the policy models. Specifically, compared to traditional GRPO, which applies sequence-level advantages, we apply token-level advantages by targeting penalties on hallucinated spans identified by Perceval, thus enabling fine-grained supervision signals. In addition to augmenting the training process, Perceval can also assist VLMs during the inference stage. Using Perceval, we can truncate the erroneous portions of the model's response, and then either have the model regenerate the response directly or induce the model to reflect on its previous output. This process can be repeated multiple times to achieve test-time scaling. Experiments show significant improvements on benchmarks from various domains across multiple reasoning VLMs trained with RL, highlighting the promise of perception-centric supervision as a general-purpose strategy. For test-time scaling, it also demonstrates consistent performance gains over other strategies, such as major voting. Our code and data will be publicly released at https://github.com/RUCAIBox/Perceval.

Yifan Li, Zhen Tan, Kai Shu et al.

Graph Neural Networks (GNNs) have emerged as a powerful tool for representation learning on graphs, but they often suffer from overfitting and label noise issues, especially when the data is scarce or imbalanced. Different from the paradigm of previous methods that rely on single-node confidence, in this paper, we introduce a novel Class-wise Selection for Graph Neural Networks, dubbed CSGNN, which employs a neighbor-aggregated latent space to adaptively select reliable nodes across different classes. Specifically, 1) to tackle the class imbalance issue, we introduce a dynamic class-wise selection mechanism, leveraging the clustering technique to identify clean nodes based on the neighbor-aggregated confidences. In this way, our approach can avoid the pitfalls of biased sampling which is common with global threshold techniques. 2) To alleviate the problem of noisy labels, built on the concept of the memorization effect, CSGNN prioritizes learning from clean nodes before noisy ones, thereby iteratively enhancing model performance while mitigating label noise. Through extensive experiments, we demonstrate that CSGNN outperforms state-of-the-art methods in terms of both effectiveness and robustness.

Yifan Li, Arravind Subramanian, Xiaoqing Liu et al.

A central challenge in computational catalysis is the identification of low-energy and chemically plausible adsorption configurations, as these directly affect adsorption energies, reaction pathways, and catalytic performance. Existing approaches generally rely on enumerating candidate adsorption sites followed by iterative refinement through density functional theory calculations or machine-learning-based relaxations. However, such workflows remain computationally expensive and are difficult to scale to complex surfaces or multi-adsorbate systems. Here, we introduce Meta-LegNet, a graph learning framework that combines SE(3)-equivariant atom-level message passing with voxel-based multiscale aggregation and cross-domain meta-learning to learn transferable representations of local adsorption environments across diverse catalyst--adsorbate systems. Rather than following a conventional regression-only paradigm, Meta-LegNet encodes local chemical environments using invariant radial features and equivariant directional information, and further incorporates broader structural context through coordinate-frame voxel pooling, assignment-based upsampling, and gated feature fusion. The resulting local-global decomposition produces atom-resolved attribution maps, which are processed to identify adsorption-relevant local environments in an interpretable manner. Based on the learned representations, we further construct an adsorption-environment database and develop a template-matching strategy to propose likely adsorption sites on previously unexplored surfaces without exhaustive site enumeration. Overall, our results suggest that learning transferable adsorption environments provides an accurate, interpretable, and practical route for accelerating catalyst screening.

Yufei Zhan, Ziheng Wu, Yousong Zhu et al.

Despite notable advancements in multimodal reasoning, leading Multimodal Large Language Models (MLLMs) still underperform on vision-centric multimodal reasoning tasks in general scenarios. This shortfall stems from their predominant reliance on logic- and knowledge-based slow thinking strategies, while effective for domains like math and science, fail to integrate visual information effectively during reasoning. Consequently, these models often fail to adequately ground visual cues, resulting in suboptimal performance in tasks that require multiple plausible visual interpretations and inferences. To address this, we present GThinker (General Thinker), a novel reasoning MLLM excelling in multimodal reasoning across general scenarios, mathematics, and science. GThinker introduces Cue-Rethinking, a flexible reasoning pattern that grounds inferences in visual cues and iteratively reinterprets these cues to resolve inconsistencies. Building on this pattern, we further propose a two-stage training pipeline, including pattern-guided cold start and incentive reinforcement learning, designed to enable multimodal reasoning capabilities across domains. Furthermore, to support the training, we construct GThinker-11K, comprising 7K high-quality, iteratively-annotated reasoning paths and 4K curated reinforcement learning samples, filling the data gap toward general multimodal reasoning. Extensive experiments demonstrate that GThinker achieves 81.5% on the challenging comprehensive multimodal reasoning benchmark M$^3$CoT, surpassing the latest O4-mini model. It also shows an average improvement of 2.1% on general scenario multimodal reasoning benchmarks, while maintaining on-par performance in mathematical reasoning compared to counterpart advanced reasoning models. The code, model, and data will be released soon at https://github.com/jefferyZhan/GThinker.

Wenhao Yu, Shaohang Wei, Jiahong Liu et al.

Token-level reweighting is a simple yet effective mechanism for controlling supervised fine-tuning, but common indicators are largely one-dimensional: the ground-truth probability reflects downstream alignment, while token entropy reflects intrinsic uncertainty induced by the pre-training prior. Ignoring entropy can misidentify noisy or easily replaceable tokens as learning-critical, while ignoring probability fails to reflect target-specific alignment. RankTuner introduces a probability--entropy calibration signal, the Relative Rank Indicator, which compares the rank of the ground-truth token with its expected rank under the prediction distribution. The inverse indicator is used as a token-wise Relative Scale to reweight the fine-tuning objective, focusing updates on truly under-learned tokens without over-penalizing intrinsically uncertain positions. Experiments on multiple backbones show consistent improvements on mathematical reasoning benchmarks, transfer gains on out-of-distribution reasoning, and pre code generation performance over probability-only or entropy-only reweighting baselines.

Tao Liu, Xutao Mao, Hongying Zan et al.

Text-to-SQL is a critical task in natural language processing that aims to transform natural language questions into accurate and executable SQL queries. In real-world scenarios, these reasoning tasks are often accompanied by complex mathematical computations, domain knowledge, and hypothetical reasoning scenarios. However, existing large-scale Text-to-SQL datasets typically focus on business logic and task logic, neglecting critical factors such as vertical domain knowledge, complex mathematical reasoning, and hypothetical reasoning, which are essential for realistically reflecting the reasoning demands in practical applications and completing data querying and analysis. To bridge this gap, we introduce LogicCat, the first Text-to-SQL benchmark dataset specifically designed for complex reasoning and chain-of-thought parsing, encompassing physics, arithmetic, commonsense, and hypothetical reasoning scenarios. LogicCat comprises 4,038 English questions paired 12,114 detailed chain-of-thought reasoning steps, spanning 45 databases across diverse domains, significantly surpassing existing datasets in complexity. Experimental results demonstrate that LogicCat substantially increases the task difficulty for current state-of-the-art models to at most 33.20% execution accuracy, indicating that this task remains exceptionally challenging. The advancement of LogicCat represents a crucial step toward developing systems suitable for real-world enterprise data analysis and autonomous query generation. We have released our dataset code at https://github.com/Ffunkytao/LogicCat.

Yifan Li, Mehrdad Salimitari, Taiyu Zhang et al.

Detection of rare lesions in whole-body CT is fundamentally limited by extreme class imbalance and low target-to-volume ratios, producing precision collapse despite high AUROC. Synthetic augmentation with diffusion models offers promise, yet pixel-space diffusion is computationally expensive, and existing mask-conditioned approaches lack controllable attribute-level regulation and paired supervision for accountable training. We introduce SALIENT, a mask-conditioned wavelet-domain diffusion framework that synthesizes paired lesion-masking volumes for controllable CT augmentation under long-tail regimes. Instead of denoising in pixel space, SALIENT performs structured diffusion over discrete wavelet coefficients, explicitly separating low-frequency brightness from high-frequency structural detail. Learnable frequency-aware objectives disentangle target and background attributes (structure, contrast, edge fidelity), enabling interpretable and stable optimization. A 3D VAE generates diverse volumetric lesion masks, and a semi-supervised teacher produces paired slice-level pseudo-labels for downstream mask-guided detection. SALIENT improves generative realism, as reflected by higher MS-SSIM (0.63 to 0.83) and lower FID (118.4 to 46.5). In a separate downstream evaluation, SALIENT-augmented training improves long-tail detection performance, yielding disproportionate AUPRC gains across low prevalences and target-to-volume ratios. Optimal synthetic ratios shift from 2x to 4x as labeled seed size decreases, indicating a seed-dependent augmentation regime under low-label conditions. SALIENT demonstrates that frequency-aware diffusion enables controllable, computationally efficient precision rescue in long-tail CT detection.

Ziqi Chen, Yi Ren, Yixuan Huang et al.

Integrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments because they lack an intrinsic understanding of electromagnetic wave propagation. We introduce the Wireless World Model (WWM), a multi-modal foundation framework predicting the spatiotemporal evolution of wireless channels by internalizing the causal relationship between 3D geometry and signal dynamics. Pre-trained on a massive ray-traced multi-modal dataset, WWM overcomes the data authenticity gap, further validated under real-world measurement data. Using a joint-embedding predictive architecture with a multi-modal mixture-of-experts Transformer, WWM fuses channel state information, 3D point clouds, and user trajectories into a unified representation. Across the five key downstream tasks supported by WWM, it achieves remarkable performance in seen environments, unseen generalization scenarios, and real-world measurements, consistently outperforming SOTA uni-modal foundation models and task-specific models. This paves the way for physics-aware 6G intelligence that adapts to the physical world.

Yifan Li, Kun Zhou, Wayne Xin Zhao et al.

As scaling up training data has significantly improved the general multimodal capabilities of Large Vision-Language Models (LVLMs), they still suffer from the hallucination issue, generating text that is inconsistent with the visual input. This phenomenon motivates us to systematically investigate the role of training data in hallucination. We introduce a new benchmark, POPEv2, which consists of counterfactual images collected from the training data of LVLMs with certain objects masked. Through comprehensive evaluation on POPEv2, we find that current LVLMs suffer from training bias: they fail to fully leverage their training data and hallucinate more frequently on images seen during training. Specifically, they perform poorly on counterfactual images, often incorrectly answering ``Yes'' to questions about masked objects. To understand this issue, we conduct probing experiments on the models' internal components, revealing that this training bias is primarily located in the language modeling (LM) head. Based on these findings, we propose Obliviate, an efficient and lightweight unlearning method designed to mitigate object hallucination via training bias unlearning. Obliviate identifies the discrepancy between ground-truth labels and model outputs on the training data as a proxy for bias and adopts a parameter- and data-efficient fine-tuning strategy that only updates the LM head. Extensive experiments demonstrate the effectiveness of our approach. While only reusing the training data and updating approximately 2\% of the parameters, Obliviate significantly reduces hallucination across both discriminative and generative tasks. Furthermore, it demonstrates strong scalability with respect to both model size (2B to 72B) and training data volume, and exhibits promising generalization to hallucination types beyond object-level hallucination. Our code and data will be publicly released.

Yifan Li, Wentao Bao, Botao Ye et al.

To effectively reduce the visual tokens in Visual Large Language Models (VLLMs), we propose a novel approach called Window Token Concatenation (WiCo). Specifically, we employ a sliding window to concatenate spatially adjacent visual tokens. However, directly concatenating these tokens may group diverse tokens into one, and thus obscure some fine details. To address this challenge, we propose fine-tuning the last few layers of the vision encoder to adaptively adjust the visual tokens, encouraging that those within the same window exhibit similar features. To further enhance the performance on fine-grained visual understanding tasks, we introduce WiCo+, which decomposes the visual tokens in later layers of the LLM. Such a design enjoys the merits of the large perception field of the LLM for fine-grained visual understanding while keeping a small number of visual tokens for efficient inference. We perform extensive experiments on both coarse- and fine-grained visual understanding tasks based on LLaVA-1.5 and Shikra, showing better performance compared with existing token reduction projectors. The code is available: https://github.com/JackYFL/WiCo.

Yifan Li, Shuai Yang, Jiaying Liu

Image colorization aims to bring colors back to grayscale images. Automatic image colorization methods, which requires no additional guidance, struggle to generate high-quality images due to color ambiguity, and provides limited user controllability. Thanks to the emergency of cross-modality datasets and models, language-based colorization methods are proposed to fully utilize the efficiency and flexibly of text descriptions to guide colorization. In view of the lack of a comprehensive review of language-based colorization literature, we conduct a thorough analysis and benchmarking. We first briefly summarize existing automatic colorization methods. Then, we focus on language-based methods and point out their core challenge on cross-modal alignment. We further divide these methods into two categories: one attempts to train a cross-modality network from scratch, while the other utilizes the pre-trained cross-modality model to establish the textual-visual correspondence. Based on the analyzed limitations of existing language-based methods, we propose a simple yet effective method based on distilled diffusion model. Extensive experiments demonstrate that our simple baseline can produces better results than previous complex methods with 14 times speed up. To the best of our knowledge, this is the first comprehensive review and benchmark on language-based image colorization field, providing meaningful insights for the community. The code is available at https://github.com/lyf1212/Color-Turbo.

Jinyi Luo, Minghao Liu, Yifan Li et al.

Super-resolution (SR) is a severely ill-posed problem with inherent ambiguity, as widely recognized in both empirical and theoretical studies. Although recent semantic-guided and multi-modal SR methods exploit large models or external priors to enhance semantic alignment, the fusion of heterogeneous modalities remains insufficiently understood in practice and theory. In this work, we provide the first theoretical modeling of multi-modal SR, revealing that prior methods are bottlenecked by sub-optimal modality utilization. Our analysis shows that the generalization risk bound can be improved by strengthening the alignment between modality weights and their effective contributions, while reducing representation complexity. This theoretical insight inspires us to propose the novel Multi-Modal Mixture-of-Experts Super-Resolution framework (M$^3$ESR) that employs generalization-oriented dynamic modality fusion for accurate risk control and modality contribution optimization. In detail, we propose a novel spatially dynamic modality weighting module and a temporally adaptive modality temperature scheduling mechanism, enabling flexible and adaptive spatial-temporal modality weighting for effective risk control. Extensive experiments demonstrate that our M$^3$ESR significantly boosts generalization and semantic consistency performances, which confirms our superiority.

Yifan Li, Shiying Wang, Jianqiang Huang

Vision-Language Pre-training (VLP) models like CLIP have significantly advanced Remote Sensing Image-Text Retrieval (RSITR). However, existing methods predominantly rely on coarse-grained global alignment, which often overlooks the dense, multi-scale semantics inherent in overhead imagery. Moreover, adapting these heavy models via full fine-tuning incurs prohibitive computational costs and risks catastrophic forgetting. To address these challenges, we propose MPS-CLIP, a parameter-efficient framework designed to shift the retrieval paradigm from global matching to keyword-guided fine-grained alignment. Specifically, we leverage a Large Language Model (LLM) to extract core semantic keywords, guiding the Segment Anything Model (SamGeo) to generate semantically relevant sub-perspectives. To efficiently adapt the frozen backbone, we introduce a Gated Global Attention (G^2A) adapter, which captures global context and long-range dependencies with minimal overhead. Furthermore, a Multi-Perspective Representation (MPR) module aggregates these local cues into robust multi-perspective embeddings. The framework is optimized via a hybrid objective combining multi-perspective contrastive and weighted triplet losses, which dynamically selects maximum-response perspectives to suppress noise and enforce precise semantic matching. Extensive experiments on the RSICD and RSITMD benchmarks demonstrate that MPS-CLIP achieves state-of-the-art performance with 35.18% and 48.40% mean Recall (mR), respectively, significantly outperforming full fine-tuning baselines and recent competitive methods. Code is available at https://github.com/Lcrucial1f/MPS-CLIP.

Haochen Luo, Zhengzhao Lai, Junjie Xu et al.

Large Language Models (LLMs) excel at general code generation, yet translating natural-language trading intents into correct option strategies remains challenging. Real-world option design requires reasoning over massive, multi-dimensional option chain data with strict constraints, which often overwhelms direct generation methods. We introduce the Option Query Language (OQL), a domain-specific intermediate representation that abstracts option markets into high-level primitives under grammatical rules, enabling LLMs to function as reliable semantic parsers rather than free-form programmers. OQL queries are then validated and executed deterministically by an engine to instantiate executable strategies. We also present a new dataset for this task and demonstrate that our neuro-symbolic pipeline significantly improves execution accuracy and logical consistency over direct baselines.

Yifan Li, Fenghe Tang, Yingtai Li et al.

General-purpose large Vision-Language Models (VLMs) demonstrate strong capabilities in generating detailed descriptions for natural images. However, their performance in the medical domain remains suboptimal, even for relatively straightforward tasks, primarily due to the lack of large-scale, high-quality, specialized medical imaging datasets and the neglect of the diagnostic process that progresses from coarse to fine-grained. To address the first issue, we construct the CT-RATE-VQA dataset, which has 84K QA pairs. For the second issue, we propose MedReason-R1, a medical VLM with explicit reasoning process for disease diagnosis. MedReason-R1 incorporates a novel strategy that embeds zoom-in disease region-of-interest areas into the image, highlighting the crucial role of both global localization and disease-specific details in enhancing the model's diagnostic performance. Furthermore, we introduce the GRPO reinforcement learning framework to MedReason-R1, which enables effective reasoning without relying on costly manual annotations. Compared to recent general-purpose and medical VLMs, MedReason-R1 achieves state-of-the-art performance in CT disease diagnosis while retaining generalization. The code, checkpoints, and dataset are available at: https://github.com/Leevan001/MedReason-R1

Yifan Li, Zhenghao Chen, Ziheng Wu et al.

Recent advances in inference-time scaling, particularly those leveraging reinforcement learning with verifiable rewards, have substantially enhanced the reasoning capabilities of Large Vision-Language Models (LVLMs). Inspired by this success, similar strategies have been applied to multimodal reasoning, yet their impact on visual perception remains unclear. To investigate this gap, we introduce DisTANCE, a perception-centric benchmark for visual estimation tasks. Evaluation results show that LVLMs exhibit limited estimation precision, and inference-time scaling offers only marginal gains. We attribute this to the fast perception paradigm of current LVLMs, where visual understanding is treated as a one-shot output without modeling the underlying perceptual process. To address this, we propose Perception-Time Scaling (PTS), a novel paradigm that encourages token-rich perception and decomposes complex perception problems into intermediate tractable sub-problems, thereby enabling perception to align with and benefit from inference-time scaling. Combined with reinforcement learning techniques, PTS significantly improves perception accuracy, raising high-precision performance on DisTANCE from 8.0% to 64.7%, and generalizes well to out-of-domain tasks. Surprisingly, even though PTS data are purely synthetic, combining them with math reasoning data yields consistent gains in both reasoning and real-world perception benchmarks. Further analysis reveals that PTS introduces more perception-related tokens and increases the model's attention to image tokens. Our code and data will be publicly released.

Dong Liu, Yanxuan Yu, Jiayi Zhang et al.

Diffusion Transformers (DiT) are powerful generative models but remain computationally intensive due to their iterative structure and deep transformer stacks. To alleviate this inefficiency, we propose FastCache, a hidden-state-level caching and compression framework that accelerates DiT inference by exploiting redundancy within the model's internal representations. FastCache introduces a dual strategy: (1) a spatial-aware token selection mechanism that adaptively filters redundant tokens based on hidden state saliency, and (2) a transformer-level cache that reuses latent activations across timesteps when changes are statistically insignificant. These modules work jointly to reduce unnecessary computation while preserving generation fidelity through learnable linear approximation. Theoretical analysis shows that FastCache maintains bounded approximation error under a hypothesis-testing-based decision rule. Empirical evaluations across multiple DiT variants demonstrate substantial reductions in latency and memory usage, with best generation output quality compared to other cache methods, as measured by FID and t-FID. Code implementation of FastCache is available on GitHub at https://github.com/NoakLiu/FastCache-xDiT.

Yifan Du, Kun Zhou, Yuqi Huo et al.

With the rapid development of video Multimodal Large Language Models (MLLMs), numerous benchmarks have been proposed to assess their video understanding capability. However, due to the lack of rich events in the videos, these datasets may suffer from the short-cut bias that the answers can be deduced from a few frames, without the need to watch the entire video. To address this issue, we introduce Event-Bench, an event-oriented long video understanding benchmark built on existing datasets and human annotations. Event-Bench includes six event-related tasks and 2,190 test instances to comprehensively evaluate video event understanding ability. Additionally, we propose Video Instruction Merging~(VIM), a cost-effective method that enhances video MLLMs using merged, event-intensive video instructions, addressing the scarcity of human-annotated, event-intensive data. Extensive experiments show that the best-performing model, GPT-4o, achieves an overall accuracy of 53.33, significantly outperforming the best open-source model by 41.42%. Leveraging an effective instruction synthesis method and an adaptive model architecture, VIM surpasses both state-of-the-art open-source models and GPT-4V on the Event-Bench. All code, data, and models are publicly available at https://github.com/RUCAIBox/Event-Bench.

Yifan Li, Hangyu Guo, Kun Zhou et al.

In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR) of 90.26% for LLaVA-1.5 and 71.60% for Gemini Pro Vision. Our code and data are available at https://github.com/RUCAIBox/HADES.

Yifan Li, Yifan Du, Kun Zhou et al.

Inspired by the superior language abilities of large language models (LLM), large vision-language models (LVLM) have been recently explored by integrating powerful LLMs for improving the performance on complex multimodal tasks. Despite the promising progress on LVLMs, we find that LVLMs suffer from the hallucination problem, i.e. they tend to generate objects that are inconsistent with the target images in the descriptions. To investigate it, this work presents the first systematic study on object hallucination of LVLMs. We conduct the evaluation experiments on several representative LVLMs, and show that they mostly suffer from severe object hallucination issue. We further discuss that the visual instructions may influence the hallucination, and find that: objects that frequently occur in the visual instructions or co-occur with the image objects, are obviously prone to be hallucinated by LVLMs. Besides, we find that existing evaluation methods might be affected by the input instructions and generation styles of LVLMs. Thus, we further design an improved evaluation method for object hallucination by proposing a polling-based query method called POPE. Experiment results demonstrate that our POPE can evaluate the object hallucination in a more stable and flexible way. Our codes and data are publicly available at https://github.com/RUCAIBox/POPE.

Mario Michael Krell, Julia Bernd, Yifan Li et al.

Social multimedia users are increasingly sharing all kinds of data about the world. They do this for their own reasons, not to provide data for field studies-but the trend presents a great opportunity for scientists. The Yahoo Flickr Creative Commons 100 Million (YFCC100M) dataset comprises 99 million images and nearly 800 thousand videos from Flickr, all shared under Creative Commons licenses. To enable scientists to leverage these media records for field studies, we propose a new framework that extracts targeted subcorpora from the YFCC100M, in a format usable by researchers who are not experts in big data retrieval and processing. This paper discusses a number of examples from the literature-as well as some entirely new ideas-of natural and social science field studies that could be piloted, supplemented, replicated, or conducted using YFCC100M data. These examples illustrate the need for a general new open-source framework for Multimedia Big Data Field Studies. There is currently a gap between the separate aspects of what multimedia researchers have shown to be possible with consumer-produced big data and the follow-through of creating a comprehensive field study framework that supports scientists across other disciplines. To bridge this gap, we must meet several challenges. For example, the framework must handle unlabeled and noisily labeled data to produce a filtered dataset for a scientist-who naturally wants it to be both as large and as clean as possible. This requires an iterative approach that provides access to statistical summaries and refines the search by constructing new classifiers. The first phase of our framework is available as Multimedia Commons Search, an intuitive interface that enables complex search queries at a large scale...

Yifan Li, Giulia Guidi

In computational science and data analytics, many workloads involve irregular and sparse computations that are inherently difficult to optimize for modern hardware. A key kernel is Sparse General Matrix-Matrix Multiplication (SpGEMM), which underpins simulations, graph analytics, and machine learning applications. SpGEMM exhibits irregular memory access patterns and workload imbalance, making it challenging to achieve high performance on GPUs. Current GPU SpGEMM solutions typically rely on a two-pass workflow to address load imbalance and reduce memory access. The symbolic pass, which determines the number of output elements per row, accounts for roughly 28% of the total runtime on average. In this work, we question the necessity of exact symbolic computation and introduce an estimation-based SpGEMM workflow. Our approach replaces the costly symbolic step with lightweight HyperLogLog estimators, combined with an analysis strategy that dynamically selects the optimal workflow and guides accumulator configuration. In addition, we introduce a hybrid accumulator design, including a novel hash-based accumulator that leverages both shared and global memory. Our approach consistently outperforms leading GPU SpGEMM implementations across a wide range of both square and rectangular matrices, achieving speedups of 1.4x-2.8x on NVIDIA A100 and H100 architectures.

Zhihan Guo, Feiyang Xu, Yifan Li et al.

The explosive growth in academic literature necessitates automated deep research (DR) agents, yet their evaluation remains a significant challenge. First, existing benchmarks often focus narrowly on retrieval while neglecting high-level planning and reasoning. Second, existing benchmarks favor general domains over the scientific domains that are the core application for DR agents. To address these gaps, we introduce Dr.Mi-Bench, a Modular-integrated benchmark for scientific DR agents. Grounded in academic literature, our benchmark uses a human-annotated dataset of 200 instances across 10 scientific domains, including both research and review papers. Besides, we also propose a Modular-integrated Evaluation Paradigm for DR Agents (Dr.Mi-Eval), a novel modular-integrated evaluation paradigm, which leverages the rich structure of academic papers to assess the core competencies of planning, retrieval, and reasoning through two complementary modes: an end-to-end evaluation for DR agents and an isolated evaluation for foundational LLMs as potential backbones. Experimental results reveal a fragmented performance landscape: agents exhibit specialized strengths but share critical weaknesses, most notably in performing the multi-source retrieval required for review-style tasks and performing consistently across diverse scientific fields. Moreover, improving high-level planning capability is the crucial factor for unlocking the reasoning potential of foundational LLMs as backbones. By exposing these actionable failure modes, Dr.Mi-Bench provides a diagnostic tool to guide the development of more reliable academic research assistants.

Hongyu Wang, Xiaotao Liu, Yifan Li et al.

RGBT tracking has been widely used in various fields such as robotics, surveillance processing, and autonomous driving. Existing RGBT trackers fully explore the spatial information between the template and the search region and locate the target based on the appearance matching results. However, these RGBT trackers have very limited exploitation of temporal information, either ignoring temporal information or exploiting it through online sampling and training. The former struggles to cope with the object state changes, while the latter neglects the correlation between spatial and temporal information. To alleviate these limitations, we propose a novel Temporal Adaptive RGBT Tracking framework, named as TATrack. TATrack has a spatio-temporal two-stream structure and captures temporal information by an online updated template, where the two-stream structure refers to the multi-modal feature extraction and cross-modal interaction for the initial template and the online update template respectively. TATrack contributes to comprehensively exploit spatio-temporal information and multi-modal information for target localization. In addition, we design a spatio-temporal interaction (STI) mechanism that bridges two branches and enables cross-modal interaction to span longer time scales. Extensive experiments on three popular RGBT tracking benchmarks show that our method achieves state-of-the-art performance, while running at real-time speed.

Yifan Li, Yingda Yin, Lingting Zhu et al.

Reasoning-centric video object segmentation is an inherently complex task: the query often refers to dynamics, causality, and temporal interactions, rather than static appearances. Yet existing solutions generally collapse these factors into simplified reasoning with latent embeddings, rendering the reasoning chain opaque and essentially intractable. We therefore adopt an explicit decomposition perspective and introduce ReVSeg, which executes reasoning as sequential decisions in the native interface of pretrained vision language models (VLMs). Rather than folding all reasoning into a single-step prediction, ReVSeg executes three explicit operations -- semantics interpretation, temporal evidence selection, and spatial grounding -- aligning pretrained capabilities. We further employ reinforcement learning to optimize the multi-step reasoning chain, enabling the model to self-refine its decision quality from outcome-driven signals. Experimental results demonstrate that ReVSeg attains state-of-the-art performances on standard video object segmentation benchmarks and yields interpretable reasoning trajectories. Project page is available at https://clementine24.github.io/ReVSeg/ .

Xinying Weng, Yifan Li, Shuaidong Hao et al.

This project proposes a new method that uses fuzzy comprehensive evaluation method to integrate ResNet-50 self-supervised and RepVGG supervised learning. The source image dataset HWOBC oracle is taken as input, the target image is selected, and finally the most similar image is output in turn without any manual intervention. The same feature encoding method is not used for images of different modalities. Before the model training, the image data is preprocessed, and the image is enhanced by random rotation processing, self-square graph equalization theory algorithm, and gamma transform, which effectively enhances the key feature learning. Finally, the fuzzy comprehensive evaluation method is used to combine the results of supervised training and unsupervised training, which can better solve the "most similar" problem that is difficult to quantify. At present, there are many unknown oracle-bone inscriptions waiting for us to crack. Contacting with the glyphs can provide new ideas for cracking.