NeurIPS 2025 Papers

Action recognition is a fundamental task in video understanding. Existing methods typically extract unified features to process all actions in one video, which makes it challenging to model the interactions between different objects in multi-action scenarios. To alleviate this issue, we explore disentangling any specified actions from complex scenes as an effective solution. In this paper, we propose Prompt-guided Disentangled Representation for Action Recognition (ProDA), a novel framework that disentangles any specified actions from a multi-action scene. ProDA leverages Spatio-temporal Scene Graphs (SSGs) and introduces Dynamic Prompt Module (DPM) to guide a Graph Parsing Neural Network (GPNN) in generating action-specific representations. Furthermore, we design a video-adapted GPNN that aggregates information using dynamic weights. Extensive experiments on two complex video action datasets, Charades and SportsHHI, demonstrate the effectiveness of our approach against state-of-the-art methods. Our code can be found in https://github.com/iamsnaping/ProDA.git.

Tabular datasets are inherently heterogeneous, presenting significant challenges for developing pre-trained foundation models. The recently introduced transformer-based Tabular Prior-data Fitted Network v2 (TabPFN v2) achieves unprecedented *in-context learning* performance across diverse downstream datasets, marking a pivotal advancement in tabular foundation models. In this paper, we take a closer look at TabPFN v2 to examine how it effectively handles heterogeneity and achieves high predictive accuracy, and to explore how its limitations in high-dimensional, many-category, and large-scale tasks can be mitigated. We find that TabPFN v2 can infer attribute relationships even when provided with randomized attribute token inputs, eliminating the need to explicitly learn dataset-specific attribute embeddings to address heterogeneity. We further show that TabPFN v2 can be transformed into a feature extractor, revealing its ability to construct a highly separable feature space for accurate predictions. Lastly, we demonstrate that TabPFN v2's limitations can be addressed through a test-time divide-and-conquer strategy, enabling scalable inference without requiring re-training. By uncovering the mechanisms behind TabPFN v2's success and introducing strategies to extend its applicability, this study offers key insights into the design of future tabular foundation models.

Existing large-scale video generation models are computationally intensive, preventing adoption in real-time and interactive applications. In this work, we propose autoregressive adversarial post-training (AAPT) to turn a pre-trained latent video diffusion model into a real-time, interactive, streaming video generator. Our model autoregressively generates a latent frame at a time using a single neural function evaluation (1NFE). The model can stream the result to the user in real time and receive interactive responses as control to generate the next latent frame. Unlike existing approaches, our method explores adversarial training as an effective paradigm for autoregressive generation. This allows us to design a more efficient architecture for one-step generation and to train the model in a student-forcing way to mitigate error accumulation. The adversarial approach also enables us to train the model for long-duration generation fully utilizing the KV cache. As a result, our 8B model achieves real-time, 24fps, nonstop, streaming video generation at 736x416 resolution on a single H100, or 1280x720 on 8xH100 up to a minute long (1440 frames).

Grounding large language models (LLMs) in domain-specific tasks like post-hoc dash-cam driving video analysis is challenging due to their general-purpose training and lack of structured inductive biases. As vision is often the sole modality available for such analysis (i.e., no LiDAR, GPS, etc.), existing video-based vision-language models (V-VLMs) struggle with spatial reasoning, causal inference, and explainability of events in the input video. To this end, we introduce iFinder, a structured semantic grounding framework that decouples perception from reasoning by translating dash-cam videos into a hierarchical, interpretable data structure for LLMs. iFinder operates as a modular, training-free pipeline that employs pretrained vision models to extract critical cues—object pose, lane positions, and object trajectories—which are hierarchically organized into frame- and video-level structures. Combined with a three-block prompting strategy, it enables step-wise, grounded reasoning for the LLM to refine a peer V-VLM's outputs and provide accurate reasoning. Evaluations on four public dash-cam video benchmarks show that iFinder's proposed grounding with domain-specific cues—especially object orientation and global context—significantly outperforms end-to-end V-VLMs on four zero-shot driving benchmarks, with up to 39% gains in accident reasoning accuracy. By grounding LLMs with driving domain-specific representations, iFinder offers a zero-shot, interpretable, and reliable alternative to end-to-end V-VLMs for post-hoc driving video understanding.

With the exponential growth of video content, aiming at localizing relevant video moments based on natural language queries, video moment retrieval (VMR) has gained significant attention. Existing weakly supervised VMR methods focus on designing various feature modeling and modal interaction modules to alleviate the reliance on precise temporal annotations. However, these methods have poor generalization capabilities on compositional queries with novel syntactic structures or vocabulary in real-world scenarios. To this end, we propose a new task: weakly supervised compositional moment retrieval (WSCMR). This task trains models using only video-query pairs without precise temporal annotations, while enabling generalization to complex compositional queries. Furthermore, a proposal-centric network (PC-Net) is proposed to tackle this challenging task. First, video and query features are extracted through frozen feature extractors, followed by modality interaction to obtain multimodal features. Second, to handle compositional queries with explicit temporal associations, a dual-granularity proposal generator decodes multimodal global and frame-level features to obtain query-relevant proposal boundaries with fine-grained temporal perception. Third, to improve the discrimination of proposal features, a proposal feature aggregator is constructed to conduct semantic alignment of frames and queries, and employ a learnable peak-aware Gaussian distributor to fit the frame weights within the proposals to derive proposal features from the video frame features. Finally, the proposal quality is assessed based on the results of reconstructing the masked query using the obtained proposal features. To further enhance the model's ability to capture semantic associations between proposals and queries, a quality margin regularizer is constructed to dynamically stratify proposals into high and low query-relevance subsets and enhance the association between queries and common elements within proposals, and suppress spurious correlations via inter-subset contrastive learning. Notably, PC-Net achieves superior performance with 54\% fewer parameters than prior works by parameter-efficient design. Experiments on Charades-CG and ActivityNet-CG demonstrate PC-Net’s ability to generalize across diverse compositional queries. Code is available at https://github.com/mingyao1120/PC-Net.

With the rapid growth of video generative models (VGMs), it is essential to develop reliable and comprehensive automatic metrics for AI-generated videos (AIGVs). Existing methods either use off-the-shelf models optimized for other tasks or rely on human assessment data to train specialized evaluators. These approaches are constrained to specific evaluation aspects and are difficult to scale with the increasing demands for finer-grained and more comprehensive evaluations. To address this issue, this work investigates the feasibility of using multimodal large language models (MLLMs) as a unified evaluator for AIGVs, leveraging their strong visual perception and language understanding capabilities. To evaluate the performance of automatic metrics in unified AIGV evaluation, we introduce a benchmark called UVE-Bench. UVE-Bench collects videos generated by state-of-the-art VGMs and provides pairwise human preference annotations across 15 evaluation aspects. Using UVE-Bench, we extensively evaluate 18 MLLMs. Our empirical results suggest that while advanced MLLMs (e.g., Qwen2VL-72B and InternVL2.5-78B) still lag behind human evaluators, they demonstrate promising ability in unified AIGV evaluation, significantly surpassing existing specialized evaluation methods. Additionally, we conduct an in-depth analysis of key design choices that impact the performance of MLLM-driven evaluators, offering valuable insights for future research on AIGV evaluation.

Recognizing the physical states of objects and their transformations within videos is crucial for structured video understanding and enabling robust real-world applications, such as robotic manipulation. However, pretrained vision-language models often struggle to capture these nuanced dynamics and their temporal context, and specialized object state recognition frameworks may not generalize to unseen actions or objects. We introduce SAGE (State-Action Graph Embeddings), a novel framework that offers a unified model of physical state transitions by decomposing states into fine-grained, language-described visual concepts that are sharable across different objects and actions. SAGE initially leverages Large Language Models to construct a State-Action Graph, which is then multimodally refined using Vision-Language Models. Extensive experiments show that our method significantly outperforms baselines, generalizes effectively to unseen objects and actions in open-world settings. SAGE improves the prior state-of-the-art by as much as 14.6% on novel state recognition with less than 5% of its inference time.

We introduce DanmakuTPPBench, a comprehensive benchmark designed to advance multi-modal Temporal Point Process (TPP) modeling in the era of Large Language Models (LLMs). While TPPs have been widely studied for modeling temporal event sequences, existing datasets are predominantly unimodal, hindering progress in models that require joint reasoning over temporal, textual, and visual information. To address this gap, DanmakuTPPBench comprises two complementary components: (1) DanmakuTPP-Events, a novel dataset derived from the Bilibili video platform, where user-generated bullet comments (Danmaku) naturally form multi-modal events annotated with precise timestamps, rich textual content, and corresponding video frames; (2) DanmakuTPP-QA, a challenging question-answering dataset constructed via a novel multi-agent pipeline powered by state-of-the-art LLMs and multi-modal LLMs (MLLMs), targeting complex temporal-textual-visual reasoning. We conduct extensive evaluations using both classical TPP models and recent MLLMs, revealing significant performance gaps and limitations in current methods’ ability to model multi-modal event dynamics. Our benchmark establishes strong baselines and calls for further integration of TPP modeling into the multi-modal language modeling landscape. Project page: https://github.com/FRENKIE-CHIANG/DanmakuTPPBench.

Conventional video outpainting methods primarily focus on maintaining coherent textures and visual consistency across frames. However, they often fail at handling dynamic scenes due to the complex motion of objects or camera movement, leading to temporal incoherence and visible flickering artifacts across frames. This is primarily because they lack instance-aware modeling to accurately separate and track individual object motions throughout the video. In this paper, we propose a novel video outpainting framework that explicitly takes shadow-object pairs into consideration to enhance the temporal and spatial consistency of instances, even when they are temporarily invisible. Specifically, we first track the shadow-object pairs across frames and predict the instances in the scene to unveil the spatial regions of invisible instances. Then, these prediction results are fed to guide the instance-aware optical flow completion to unveil the temporal motion of invisible instances. Next, these spatiotemporal guidances of instances are used to guide the video outpainting process. Finally, a video-aware discriminator is implemented to enhance alignment among dynamic shadows and the extended semantics in the scene. Comprehensive experiments underscore the superiority of our approach, outperforming existing state-of-the-art methods in widely recognized benchmarks.

While recent multimodal large language models (MLLMs) have advanced automated ECG interpretation, they still face two key limitations: (1) insufficient multimodal synergy between ECG time series and ECG images, and (2) limited explainability in linking diagnoses to granular waveform evidence. We introduce GEM, the first MLLM unifying ECG time series, 12-lead ECG images and text for grounded and clinician-aligned ECG interpretation. GEM enables feature-grounded analysis, evidence-driven reasoning, and a clinician-like diagnostic process through three core innovations: a dual-encoder framework extracting complementary time series and image features, cross-modal alignment for effective multimodal understanding, and knowledge-guided instruction data generation for generating high-granularity grounding data (ECG-Grounding) linking diagnoses to measurable parameters ($e.g.$, QRS/PR Intervals). Additionally, we propose the Grounded ECG Understanding task, a clinically motivated benchmark designed to comprehensively assess the MLLM's capability in grounded ECG understanding. Experimental results on both existing and our proposed benchmarks show GEM significantly improves predictive performance (CSN $7.4\%$ $\uparrow$), explainability ($22.7\%$ $\uparrow$), and grounding ($25.3\%$ $\uparrow$), making it a promising approach for real-world clinical applications. Codes, model, and data are available at https://github.com/lanxiang1017/GEM.

Multimodal large language models (MLLMs) demonstrates remarkable capabilities in handling complex multimodal tasks and are increasingly adopted in video understanding applications. However, their rapid advancement raises serious data privacy concerns, particularly given the potential inclusion of sensitive video content, such as personal recordings and surveillance footage, in their training datasets. Determining improperly used videos during training remains a critical and unresolved challenge. Despite considerable progress on membership inference attacks (MIAs) for text and image data in MLLMs, existing methods fail to generalize effectively to the video domain. These methods suffer from poor scalability as more frames are sampled and generally achieve negligible true positive rates at low false positive rates (TPR@Low FPR), mainly due to their failure to capture the inherent temporal variations of video frames and to account for model behavior differences as the number of frames varies. To address these challenges, we introduce Vid-SME (**Vid**eo **S**harma–**M**ittal **E**ntropy), the first membership inference method tailored for video data used in video understanding LLMs (VULLMs). Vid-SME leverages the confidence of model output and integrates adaptive parameterization to compute Sharma–Mittal entropy (SME) for video inputs. By leveraging the SME difference between natural and temporally-reversed video frames, Vid-SME derives robust membership scores to determine whether a given video is part of the model's training set. Experiments on various self-trained and open-sourced VULLMs demonstrate the strong effectiveness of Vid-SME.

Video inpainting and editing have long been challenging tasks in the video generation community, requiring extensive computational resources and large datasets to train models with satisfactory performance. Recent breakthroughs in large-scale video foundation models have greatly enhanced text-to-video generation capabilities. This naturally leads to the idea of leveraging the prior knowledge from these powerful generators to facilitate video inpainting and editing. In this work, we investigate the feasibility of employing pre-trained text-to-video foundation models for high-quality video inpainting and editing without additional training. Specifically, we introduce a model-agnostic denoising sampler that optimizes the trajectory by maximizing the log-likelihood expectation conditioned on the known video segments. To enable efficient dynamic object removal and replacement, we propose a latent mask fuser that performs accurate video masking directly in latent space, eliminating the need for explicit VAE decoding and encoding. We implement our approach in widely-used foundation generators such as CogVideoX and HunyuanVideo, demonstrating the model-agnostic nature of our sampler. Comprehensive quantitative and qualitative evaluations confirm that our method achieves outstanding video inpainting and editing performance in a plug-and-play fashion.

Coral reefs are vital yet vulnerable ecosystems that require continuous monitoring to support conservation. While coral reef images provide essential information in coral monitoring, interpreting such images remains challenging due to the need for domain expertise. Visual Question Answering (VQA), powered by Large Vision-Language Models (LVLMs), has great potential in user-friendly interaction with coral reef images. However, applying VQA to coral imagery demands a dedicated dataset that addresses two key challenges: domain-specific annotations and multidimensional questions. In this work, we introduce CoralVQA, the first large-scale VQA dataset for coral reef analysis. It contains 12,805 real-world coral images from 67 coral genera collected from 3 oceans, along with 277,653 question-answer pairs that comprehensively assess ecological and health-related conditions. To construct this dataset, we develop a semi-automatic data construction pipeline in collaboration with marine biologists to ensure both scalability and professional-grade data quality. CoralVQA presents novel challenges and provides a comprehensive benchmark for studying vision-language reasoning in the context of coral reef images. By evaluating several state-of-the-art LVLMs, we reveal key limitations and opportunities. These insights form a foundation for future LVLM development, with a particular emphasis on supporting coral conservation efforts.

Masked-based autoregressive models have demonstrated promising image generation capability in continuous space. However, their potential for video generation remains under-explored. Masked-based autoregressive models have demonstrated promising image generation capability in continuous space. However, their potential for video generation remains under-explored. In this paper, we propose \textbf{VideoMAR}, a concise and efficient decoder-only autoregressive image-to-video model with continuous tokens, composing temporal frame-by-frame and spatial masked generation. We first identify temporal causality and spatial bi-directionality as the first principle of video AR models, and propose the next-frame diffusion loss for the integration of mask and video generation. Besides, the huge cost and difficulty of long sequence autoregressive modeling is a basic but crucial issue. To this end, we propose the temporal short-to-long curriculum learning and spatial progressive resolution training, and employ progressive temperature strategy at inference time to mitigate the accumulation error. Furthermore, VideoMAR replicates several unique capacities of language models to video generation. It inherently bears high efficiency due to simultaneous temporal-wise KV cache and spatial-wise parallel generation, and presents the capacity of spatial and temporal extrapolation via 3D rotary embeddings. On the VBench-I2V benchmark, VideoMAR surpasses the previous state-of-the-art (Cosmos I2V) while requiring significantly fewer parameters ($9.3\%$), training data ($0.5\%$), and GPU resources ($0.2\%$).

Partial observability is a notorious challenge in reinforcement learning (RL), due to the need to learn complex, history-dependent policies. Recent empirical successes have used *privileged expert distillation* -- which leverages availability of latent state information during training (e.g., from a simulator) to learn and imitate the optimal latent, Markovian policy -- to disentangle the task of ''learning to see'' from ''learning to act''. While expert distillation is more computationally efficient than RL without latent state information, it also has well-documented failure modes. In this paper -- through a simple but instructive theoretical model called the *perturbed Block MDP*, and controlled experiments on challenging simulated locomotion tasks -- we investigate the algorithmic trade-off between privileged expert distillation and standard RL without privileged information. Our main findings are: **(1)** The trade-off empirically hinges on the *stochasticity* of the latent dynamics, as theoretically predicted by contrasting *approximate decodability* with *belief contraction* in the perturbed Block MDP; and **(2)** The optimal latent policy is not always the best latent policy to distill. Our results suggest new guidelines for effectively exploiting privileged information, potentially advancing the efficiency of policy learning across many practical partially observable domains.

Our objective is to develop compact video representations that are sensitive to visual change over time. To measure such time-sensitivity, we introduce a new task: chiral action recognition, where one needs to distinguish between a pair of temporally opposite actions, such as “opening vs. closing a door", “approaching vs. moving away from something", “folding vs. unfolding paper", etc. Such actions (i) occur frequently in everyday life, (ii) require understanding of simple visual change over time (in object state, size, spatial position, count . . . ), and (iii) are known to be poorly represented by many video embeddings. Our goal is to build time aware video representations which offer linear separability between these chiral pairs. To that end, we propose a self-supervised adaptation recipe to inject time-sensitivity into a sequence of frozen image features. Our model is based on an auto-encoder with a latent space with inductive bias inspired by perceptual straightening. We show that this results in a compact but time-sensitive video representation for the proposed task across three datasets: Something-Something, EPIC-Kitchens, and Charade. Our method (i) outperforms much larger video models pre-trained on large-scale video datasets, and (ii) leads to an improvement in classification performance on standard benchmarks when combined with these existing models.

Multimodal Large Language Models (MLLMs) have achieved considerable accuracy in Optical Character Recognition (OCR) from static images. However, their efficacy in video OCR is significantly diminished due to factors such as motion blur, temporal variations, and visual effects inherent in video content. To provide clearer guidance for training practical MLLMs, we introduce MME-VideoOCR benchmark, which encompasses a comprehensive range of video OCR application scenarios. MME-VideoOCR features 10 task categories comprising 25 individual tasks and spans 44 diverse scenarios. These tasks extend beyond text recognition to incorporate deeper comprehension and reasoning of textual content within videos. The benchmark consists of 1,464 videos with varying resolutions, aspect ratios, and durations, along with 2,000 meticulously curated, manually annotated question-answer pairs. We evaluate 18 state-of-the-art MLLMs on MME-VideoOCR, revealing that even the best-performing model (Gemini-2.5 Pro) achieves only an accuracy of 73.7%. Fine-grained analysis indicates that while existing MLLMs demonstrate strong performance on tasks where relevant texts are contained within a single or few frames, they exhibit limited capability in effectively handling tasks that demand holistic video comprehension. These limitations are especially evident in scenarios that require spatio-temporal reasoning, cross-frame information integration, or resistance to language prior bias. Our findings also highlight the importance of high-resolution visual input and sufficient temporal coverage for reliable OCR in dynamic video scenarios.

Modern multimodal large language models (MLLMs) can reason over hour-long video, yet their key–value (KV) cache grows linearly with time—quickly exceeding the fixed memory of phones, AR glasses, and edge robots. Prior compression schemes either assume the whole video and user query are available offline or must first build the full cache, so memory still scales with stream length. InfiniPot-V is the first training-free, query-agnostic framework that enforces a hard, length-independent memory cap for \textit{streaming} video understanding. During video encoding it monitors the cache and, once a user-set threshold is reached, runs a lightweight compression pass that (i) removes temporally redundant tokens via Temporal-axis Redundancy (TaR) metric and (ii) keeps semantically significant tokens via Value-Norm (VaN) ranking. Across four open-source MLLMs and four long-video and streaming-video benchmarks, InfiniPot-V cuts peak GPU memory by up to 94\%, sustains real-time generation, and matches or surpasses full-cache accuracy—even in multi-turn dialogues. By dissolving the KV cache bottleneck without retraining or query knowledge, InfiniPot-V closes the gap for on-device streaming video assistants.

Given a task and a set of steps composing it, Video Step Grounding (VSG) aims to detect which steps are performed in a video. Standard approaches for this task require a labeled training set (e.g., with step-level annotations or narrations), which may be costly to collect. Moreover, they process the full video offline, limiting their applications for scenarios requiring online decisions. Thus, in this work, we explore how to perform VSG online and without training. We achieve this by exploiting the zero-shot capabilities of recent Large Multimodal Models (LMMs). In particular, we use LMMs to predict the step associated with a restricted set of frames, without access to the whole video. We show that this online strategy without task-specific tuning outperforms offline and training-based models. Motivated by this finding, we develop Bayesian Grounding with Large Multimodal Models (BaGLM), further injecting knowledge of past frames into the LMM-based predictions. BaGLM exploits Bayesian filtering principles, modeling step transitions via (i) a dependency matrix extracted through large language models and (ii) an estimation of step progress. Experiments on three datasets show superior performance of BaGLM over state-of-the-art training-based offline methods.