Mingfang Zhang

Eshika Khandelwal, Jingjing Pan, Mingfang Zhang et al.

Multimodal large language models (MLLMs) are predominantly evaluated on free-form vision-language tasks such as visual question answering, captioning, and summarization. However, their practical use is rapidly expanding to more structured computer vision settings, where users prompt models to perform localization-centric tasks such as object detection, often within larger agentic or decision-making systems. Despite this shift, there is currently no standardized benchmark that systematically evaluates these capabilities at scale. In this work, we introduce the first comprehensive benchmark specifically designed to assess the promptable localization abilities of generalist MLLMs. Our benchmark spans four core task categories: object detection, referring expression detection, instance-level detection, and video-based detection. To enable consistent and fair evaluation, we develop a unified framework that standardizes inputs, enforces parsable bounding box outputs, and defines transparent evaluation protocols across tasks. Using this suite, we evaluate a diverse set of open-source and proprietary MLLMs, providing an in-depth analysis of their performance and limitations. Beyond accuracy, we examine models' ability to adhere to output format specifications, showing that current systems are highly sensitive to formatting constraints and often fail to generalize even to minor variations. Our results highlight both the strengths and shortcomings of state-of-the-art MLLMs in localization settings, and point toward important directions for improving multimodal model design and evaluation.

Mingfang Zhang, Yunfei Liu, Feng Lu

Appearance-based gaze estimation aims to predict the 3D eye gaze direction from a single image. While recent deep learning-based approaches have demonstrated excellent performance, they usually assume one calibrated face in each input image and cannot output multi-person gaze in real time. However, simultaneous gaze estimation for multiple people in the wild is necessary for real-world applications. In this paper, we propose the first one-stage end-to-end gaze estimation method, GazeOnce, which is capable of simultaneously predicting gaze directions for multiple faces (>10) in an image. In addition, we design a sophisticated data generation pipeline and propose a new dataset, MPSGaze, which contains full images of multiple people with 3D gaze ground truth. Experimental results demonstrate that our unified framework not only offers a faster speed, but also provides a lower gaze estimation error compared with state-of-the-art methods. This technique can be useful in real-time applications with multiple users.

Mingfang Zhang, Yifei Huang, Ruicong Liu et al.

Compared with visual signals, Inertial Measurement Units (IMUs) placed on human limbs can capture accurate motion signals while being robust to lighting variation and occlusion. While these characteristics are intuitively valuable to help egocentric action recognition, the potential of IMUs remains under-explored. In this work, we present a novel method for action recognition that integrates motion data from body-worn IMUs with egocentric video. Due to the scarcity of labeled multimodal data, we design an MAE-based self-supervised pretraining method, obtaining strong multi-modal representations via modeling the natural correlation between visual and motion signals. To model the complex relation of multiple IMU devices placed across the body, we exploit the collaborative dynamics in multiple IMU devices and propose to embed the relative motion features of human joints into a graph structure. Experiments show our method can achieve state-of-the-art performance on multiple public datasets. The effectiveness of our MAE-based pretraining and graph-based IMU modeling are further validated by experiments in more challenging scenarios, including partially missing IMU devices and video quality corruption, promoting more flexible usages in the real world.

Mingfang Zhang, Jinglu Wang, Xiao Li et al.

The Multiplane Image (MPI), containing a set of fronto-parallel RGBA layers, is an effective and efficient representation for view synthesis from sparse inputs. Yet, its fixed structure limits the performance, especially for surfaces imaged at oblique angles. We introduce the Structural MPI (S-MPI), where the plane structure approximates 3D scenes concisely. Conveying RGBA contexts with geometrically-faithful structures, the S-MPI directly bridges view synthesis and 3D reconstruction. It can not only overcome the critical limitations of MPI, i.e., discretization artifacts from sloped surfaces and abuse of redundant layers, and can also acquire planar 3D reconstruction. Despite the intuition and demand of applying S-MPI, great challenges are introduced, e.g., high-fidelity approximation for both RGBA layers and plane poses, multi-view consistency, non-planar regions modeling, and efficient rendering with intersected planes. Accordingly, we propose a transformer-based network based on a segmentation model. It predicts compact and expressive S-MPI layers with their corresponding masks, poses, and RGBA contexts. Non-planar regions are inclusively handled as a special case in our unified framework. Multi-view consistency is ensured by sharing global proxy embeddings, which encode plane-level features covering the complete 3D scenes with aligned coordinates. Intensive experiments show that our method outperforms both previous state-of-the-art MPI-based view synthesis methods and planar reconstruction methods.

Nie Lin, Takehiko Ohkawa, Mingfang Zhang et al.

We present a contrastive learning framework based on in-the-wild hand images tailored for pre-training 3D hand pose estimators, dubbed HandCLR. Pre-training on large-scale images achieves promising results in various tasks, but prior 3D hand pose pre-training methods have not fully utilized the potential of diverse hand images accessible from in-the-wild videos. To facilitate scalable pre-training, we first prepare an extensive pool of hand images from in-the-wild videos and design our method with contrastive learning. Specifically, we collected over 2.0M hand images from recent human-centric videos, such as 100DOH and Ego4D. To extract discriminative information from these images, we focus on the similarity of hands; pairs of similar hand poses originating from different samples, and propose a novel contrastive learning method that embeds similar hand pairs closer in the latent space. Our experiments demonstrate that our method outperforms conventional contrastive learning approaches that produce positive pairs sorely from a single image with data augmentation. We achieve significant improvements over the state-of-the-art method in various datasets, with gains of 15% on FreiHand, 10% on DexYCB, and 4% on AssemblyHands.

Yifei Huang, Guo Chen, Jilan Xu et al.

Being able to map the activities of others into one's own point of view is one fundamental human skill even from a very early age. Taking a step toward understanding this human ability, we introduce EgoExoLearn, a large-scale dataset that emulates the human demonstration following process, in which individuals record egocentric videos as they execute tasks guided by demonstration videos. Focusing on the potential applications in daily assistance and professional support, EgoExoLearn contains egocentric and demonstration video data spanning 120 hours captured in daily life scenarios and specialized laboratories. Along with the videos we record high-quality gaze data and provide detailed multimodal annotations, formulating a playground for modeling the human ability to bridge asynchronous procedural actions from different viewpoints. To this end, we present benchmarks such as cross-view association, cross-view action planning, and cross-view referenced skill assessment, along with detailed analysis. We expect EgoExoLearn can serve as an important resource for bridging the actions across views, thus paving the way for creating AI agents capable of seamlessly learning by observing humans in the real world. Code and data can be found at: https://github.com/OpenGVLab/EgoExoLearn

Mingfang Zhang, Jingjing Pan, Ashutosh Kumar et al.

Cause-and-effect reasoning in video is a significant challenge for Vision-Language Models (VLMs), as it requires going beyond surface-level perception to a deeper understanding of causal mechanisms. However, existing benchmarks rarely provide the fine-grained, grounded evidence needed to rigorously evaluate this capability. To address this gap, we introduce CaST-Bench, a benchmark for Causal Chain-Grounded Spatio-Temporal Video Reasoning. CaST-Bench presents complex causal questions that require models to identify and localize a chain of multiple spatio-temporal evidences. Through a human-AI collaborative pipeline, we construct a high-quality dataset of 2,066 questions over 1,015 videos, with causal chains annotated by temporal segments and bounding-box tracks. Furthermore, we design a comprehensive evaluation suite with novel metrics that assess not only answer correctness but also the capability for visual evidence grounded reasoning. This grounding is crucial for improving accuracy by mitigating spurious correlations and for enhancing user trust by making models more transparent. Our experiments show that current VLMs struggle with causal questions, largely due to their limited ability to construct precise and grounded causal chains. This highlights an important direction for improving future VLMs.

Caixin Kang, Tianyu Yan, Sitong Gong et al.

Multimodal Large Language Models (MLLMs) are increasingly deployed in human-facing roles where personality perception is critical, yet existing benchmarks evaluate this capability solely on numerical Big Five score prediction, leaving open whether models truly perceive personality through behavioral understanding or merely prejudge through superficial pattern matching. We address this gap with three contributions. (i) A new task: we formalize Grounded Personality Reasoning (GPR), which requires MLLMs to anchor each Big Five rating in observable evidence through a chain of rating, reasoning, and grounding. (ii) A new dataset: we release MM-OCEAN (1,104 videos, 5,320 MCQs), produced by a multi-agent pipeline with human verification, with timestamped behavioral observations, evidence-grounded trait analyses, and seven categories of cue-grounding MCQs. (iii) Benchmark and analysis: we design a three-tier evaluation (rating, reasoning, grounding) plus four sample-level failure-mode metrics: Prejudice Rate (PR), Confabulation Rate (CR), Integration-failure Rate (IR), and Holistic-grounding Rate (HR), and benchmark 27 MLLMs (13 closed, 14 open). The analysis uncovers a striking Prejudice Gap: across the field, 51% of correct ratings are not grounded in retrieved cues, and the Holistic-Grounding Rate spans only 0-33.5%. These findings expose a disconnect between getting the right score and reasoning for the right reason, charting a roadmap for grounded social cognition in MLLMs.

Caixin Kang, Yifei Huang, Liangyang Ouyang et al.

As AI systems become increasingly integrated into human lives, endowing them with robust social intelligence has emerged as a critical frontier. A key aspect of this intelligence is discerning truth from deception, a ubiquitous element of human interaction that is conveyed through a complex interplay of verbal language and non-verbal visual cues. However, automatic deception detection in dynamic, multi-party conversations remains a significant challenge. The recent rise of powerful Multimodal Large Language Models (MLLMs), with their impressive abilities in visual and textual understanding, makes them natural candidates for this task. Consequently, their capabilities in this crucial domain are mostly unquantified. To address this gap, we introduce a new task, Multimodal Interactive Veracity Assessment (MIVA), and present a novel multimodal dataset derived from the social deduction game Werewolf. This dataset provides synchronized video, text, with verifiable ground-truth labels for every statement. We establish a comprehensive benchmark evaluating state-of-the-art MLLMs, revealing a significant performance gap: even powerful models like GPT-4o struggle to distinguish truth from falsehood reliably. Our analysis of failure modes indicates that these models fail to ground language in visual social cues effectively and may be overly conservative in their alignment, highlighting the urgent need for novel approaches to building more perceptive and trustworthy AI systems.

Ruicong Liu, Takehiko Ohkawa, Mingfang Zhang et al.

The pursuit of accurate 3D hand pose estimation stands as a keystone for understanding human activity in the realm of egocentric vision. The majority of existing estimation methods still rely on single-view images as input, leading to potential limitations, e.g., limited field-of-view and ambiguity in depth. To address these problems, adding another camera to better capture the shape of hands is a practical direction. However, existing multi-view hand pose estimation methods suffer from two main drawbacks: 1) Requiring multi-view annotations for training, which are expensive. 2) During testing, the model becomes inapplicable if camera parameters/layout are not the same as those used in training. In this paper, we propose a novel Single-to-Dual-view adaptation (S2DHand) solution that adapts a pre-trained single-view estimator to dual views. Compared with existing multi-view training methods, 1) our adaptation process is unsupervised, eliminating the need for multi-view annotation. 2) Moreover, our method can handle arbitrary dual-view pairs with unknown camera parameters, making the model applicable to diverse camera settings. Specifically, S2DHand is built on certain stereo constraints, including pair-wise cross-view consensus and invariance of transformation between both views. These two stereo constraints are used in a complementary manner to generate pseudo-labels, allowing reliable adaptation. Evaluation results reveal that S2DHand achieves significant improvements on arbitrary camera pairs under both in-dataset and cross-dataset settings, and outperforms existing adaptation methods with leading performance. Project page: https://github.com/MickeyLLG/S2DHand.

Yifei Huang, Jilan Xu, Baoqi Pei et al.

We present Vinci, a vision-language system designed to provide real-time, comprehensive AI assistance on portable devices. At its core, Vinci leverages EgoVideo-VL, a novel model that integrates an egocentric vision foundation model with a large language model (LLM), enabling advanced functionalities such as scene understanding, temporal grounding, video summarization, and future planning. To enhance its utility, Vinci incorporates a memory module for processing long video streams in real time while retaining contextual history, a generation module for producing visual action demonstrations, and a retrieval module that bridges egocentric and third-person perspectives to provide relevant how-to videos for skill acquisition. Unlike existing systems that often depend on specialized hardware, Vinci is hardware-agnostic, supporting deployment across a wide range of devices, including smartphones and wearable cameras. In our experiments, we first demonstrate the superior performance of EgoVideo-VL on multiple public benchmarks, showcasing its vision-language reasoning and contextual understanding capabilities. We then conduct a series of user studies to evaluate the real-world effectiveness of Vinci, highlighting its adaptability and usability in diverse scenarios. We hope Vinci can establish a new framework for portable, real-time egocentric AI systems, empowering users with contextual and actionable insights. Including the frontend, backend, and models, all codes of Vinci are available at https://github.com/OpenGVLab/vinci.

Nie Lin, Takehiko Ohkawa, Yifei Huang et al.

We present a framework for pre-training of 3D hand pose estimation from in-the-wild hand images sharing with similar hand characteristics, dubbed SimHand. Pre-training with large-scale images achieves promising results in various tasks, but prior methods for 3D hand pose pre-training have not fully utilized the potential of diverse hand images accessible from in-the-wild videos. To facilitate scalable pre-training, we first prepare an extensive pool of hand images from in-the-wild videos and design our pre-training method with contrastive learning. Specifically, we collect over 2.0M hand images from recent human-centric videos, such as 100DOH and Ego4D. To extract discriminative information from these images, we focus on the similarity of hands: pairs of non-identical samples with similar hand poses. We then propose a novel contrastive learning method that embeds similar hand pairs closer in the feature space. Our method not only learns from similar samples but also adaptively weights the contrastive learning loss based on inter-sample distance, leading to additional performance gains. Our experiments demonstrate that our method outperforms conventional contrastive learning approaches that produce positive pairs sorely from a single image with data augmentation. We achieve significant improvements over the state-of-the-art method (PeCLR) in various datasets, with gains of 15% on FreiHand, 10% on DexYCB, and 4% on AssemblyHands. Our code is available at https://github.com/ut-vision/SiMHand.

Long Phan, Alice Gatti, Ziwen Han et al. · amazon-science, apple-ml

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai.

Liangyang Ouyang, Yifei Huang, Mingfang Zhang et al.

Understanding social interaction in video requires reasoning over a dynamic interplay of verbal and non-verbal cues: who is speaking, to whom, and with what gaze or gestures. While Multimodal Large Language Models (MLLMs) are natural candidates, simply adding visual inputs yields surprisingly inconsistent gains on social tasks. Our quantitative analysis of cross-modal attention inside state-of-the-art MLLMs reveals a core failure mode: in multi-speaker scenes, visual and textual tokens lack speaker-consistent alignment, exhibiting substantially weaker cross-modal attention than in object-centric images. To address this, we propose a multimodal multi-speaker attention alignment method that can be integrated into existing MLLMs. First, we introduce dynamic cross-modal head selection to identify attention heads most responsible for grounding. Then, an adaptive social-aware attention bias, computed from existing attention patterns and speaker locations, is injected into the attention mechanism. This bias reinforces alignment between a speaker's visual representation and their utterances without introducing trainable parameters or architectural changes. We integrate our method into three distinct MLLMs (LLaVA-NeXT-Video, Qwen2.5-VL, and InternVL3) and evaluate on three benchmarks (TVQA+, MMSI, OnlineMMSI). Across four social tasks, results demonstrate that our approach improves the ability of MLLMs and achieves state-of-the-art results. Attention visualizations confirm our method successfully focuses the model on speaker-relevant regions, enabling more robust multi-party social reasoning. Our implementation and model will be available at https://github.com/ut-vision/SocialInteraction.

Jarod Lévy, Mingfang Zhang, Svetlana Pinet et al.

Modern neuroprostheses can now restore communication in patients who have lost the ability to speak or move. However, these invasive devices entail risks inherent to neurosurgery. Here, we introduce a non-invasive method to decode the production of sentences from brain activity and demonstrate its efficacy in a cohort of 35 healthy volunteers. For this, we present Brain2Qwerty, a new deep learning architecture trained to decode sentences from either electro- (EEG) or magneto-encephalography (MEG), while participants typed briefly memorized sentences on a QWERTY keyboard. With MEG, Brain2Qwerty reaches, on average, a character-error-rate (CER) of 32% and substantially outperforms EEG (CER: 67%). For the best participants, the model achieves a CER of 19%, and can perfectly decode a variety of sentences outside of the training set. While error analyses suggest that decoding depends on motor processes, the analysis of typographical errors suggests that it also involves higher-level cognitive factors. Overall, these results narrow the gap between invasive and non-invasive methods and thus open the path for developing safe brain-computer interfaces for non-communicating patients.

Ashutosh Kumar, Rajat Saini, Jingjing Pan et al.

Current vision-language pre-training (VLP) paradigms excel at global scene understanding but struggle with instance-level reasoning due to global-only supervision. We introduce InstAP, an Instance-Aware Pre-training framework that jointly optimizes global vision-text alignment and fine-grained, instance-level contrastive alignment by grounding textual mentions to specific spatial-temporal regions. To support this, we present InstVL, a large-scale dataset (2 million images, 50,000 videos) with dual-granularity annotations: holistic scene captions and dense, grounded instance descriptions. On the InstVL benchmark, InstAP substantially outperforms existing VLP models on instance-level retrieval, and also surpasses a strong VLP baseline trained on the exact same data corpus, isolating the benefit of our instance-aware objective. Moreover, instance-centric pre-training improves global understanding: InstAP achieves competitive zero-shot performance on multiple video benchmarks, including MSR-VTT and DiDeMo. Qualitative visualizations further show that InstAP localizes textual mentions to the correct instances, while global-only models exhibit more diffuse, scene-level attention.

Caixin Kang, Yifei Huang, Liangyang Ouyang et al.

Despite their advanced reasoning capabilities, state-of-the-art Multimodal Large Language Models (MLLMs) demonstrably lack a core component of human intelligence: the ability to `read the room' and assess deception in complex social interactions. To rigorously quantify this failure, we introduce a new task, Multimodal Interactive Deception Assessment (MIDA), and present a novel multimodal dataset providing synchronized video and text with verifiable ground-truth labels for every statement. We establish a comprehensive benchmark evaluating 12 state-of-the-art open- and closed-source MLLMs, revealing a significant performance gap: even powerful models like GPT-4o struggle to distinguish truth from falsehood reliably. Our analysis of failure modes indicates that these models fail to effectively ground language in multimodal social cues and lack the ability to model what others know, believe, or intend, highlighting the urgent need for novel approaches to building more perceptive and trustworthy AI systems. To take a step forward, we design a Social Chain-of-Thought (SoCoT) reasoning pipeline and a Dynamic Social Epistemic Memory (DSEM) module. Our framework yields performance improvement on this challenging task, demonstrating a promising new path toward building MLLMs capable of genuine human-like social reasoning.

Ruicong Liu, Takehiko Ohkawa, Tze Ho Elden Tse et al.

This paper presents RPEP, the first pre-training method for event-based 3D hand pose estimation using labeled RGB images and unpaired, unlabeled event data. Event data offer significant benefits such as high temporal resolution and low latency, but their application to hand pose estimation is still limited by the scarcity of labeled training data. To address this, we repurpose real RGB datasets to train event-based estimators. This is done by constructing pseudo-event-RGB pairs, where event data is generated and aligned with the ground-truth poses of RGB images. Unfortunately, existing pseudo-event generation techniques assume stationary objects, thus struggling to handle non-stationary, dynamically moving hands. To overcome this, RPEP introduces a novel generation strategy that decomposes hand movements into smaller, step-by-step motions. This decomposition allows our method to capture temporal changes in articulation, constructing more realistic event data for a moving hand. Additionally, RPEP imposes a motion reversal constraint, regularizing event generation using reversed motion. Extensive experiments show that our pre-trained model significantly outperforms state-of-the-art methods on real event data, achieving up to 24% improvement on EvRealHands. Moreover, it delivers strong performance with minimal labeled samples for fine-tuning, making it well-suited for practical deployment.

Mingfang Zhang, Ryo Yonetani, Yifei Huang et al.

This paper presents a novel inertial localization framework named Egocentric Action-aware Inertial Localization (EAIL), which leverages egocentric action cues from head-mounted IMU signals to localize the target individual within a 3D point cloud. Human inertial localization is challenging due to IMU sensor noise that causes trajectory drift over time. The diversity of human actions further complicates IMU signal processing by introducing various motion patterns. Nevertheless, we observe that some actions captured by the head-mounted IMU correlate with spatial environmental structures (e.g., bending down to look inside an oven, washing dishes next to a sink), thereby serving as spatial anchors to compensate for the localization drift. The proposed EAIL framework learns such correlations via hierarchical multi-modal alignment with vision-language guidance. By assuming that the 3D point cloud of the environment is available, it contrastively learns modality encoders that align short-term egocentric action cues in IMU signals with local environmental features in the point cloud. The learning process is enhanced using concurrently collected vision and language signals to improve multimodal alignment. The learned encoders are then used in reasoning the IMU data and the point cloud over time and space to perform inertial localization. Interestingly, these encoders can further be utilized to recognize the corresponding sequence of actions as a by-product. Extensive experiments demonstrate the effectiveness of the proposed framework over state-of-the-art inertial localization and inertial action recognition baselines.