SlowFocus: Enhancing Fine-grained Temporal Understanding in Video LLM

Large language models (LLMs) have demonstrated exceptional capabilities in text understanding, which has paved the way for their expansion into video LLMs (Vid-LLMs) to analyze video data. However, current Vid-LLMs struggle to simultaneously retain high-quality frame-level semantic information (i.e., a sufficient number of tokens per frame) and comprehensive video-level temporal information (i.e., an adequate number of sampled frames per video). This limitation hinders the advancement of Vid-LLMs towards fine-grained video understanding. To address this issue, we introduce the SlowFocus mechanism, which significantly enhances the equivalent sampling frequency without compromising the quality of frame-level visual tokens. SlowFocus begins by identifying the query-related temporal segment based on the posed question, then performs dense sampling on this segment to extract local high-frequency features. A multi-frequency mixing attention module is further leveraged to aggregate these local high-frequency details with global low-frequency contexts for enhanced temporal comprehension. Additionally, to tailor Vid-LLMs to this innovative mechanism, we introduce a set of training strategies aimed at bolstering both temporal grounding and detailed temporal reasoning capabilities. Furthermore, we establish FineAction-CGR, a benchmark specifically devised to assess the ability of Vid-LLMs to process fine-grained temporal understanding tasks. Comprehensive experiments demonstrate the superiority of our mechanism across both existing public video understanding benchmarks and our proposed FineAction-CGR.

Multi-Objective Optimization for Synthetic-to-Real Style Transfer

Semantic segmentation networks require large amounts of pixel-level annotated data, which are costly to obtain for real-world images. Computer graphics engines can generate synthetic images alongside their ground-truth annotations. However, models trained on such images can perform poorly on real images due to the domain gap between real and synthetic images. Style transfer methods can reduce this difference by applying a realistic style to synthetic images. Choosing effective data transformations and their sequence is difficult due to the large combinatorial search space of style transfer operators. Using multi-objective genetic algorithms, we optimize pipelines to balance structural coherence and style similarity to target domains. We study the use of paired-image metrics on individual image samples during evolution to enable rapid pipeline evaluation, as opposed to standard distributional metrics that require the generation of many images. After optimization, we evaluate the resulting Pareto front using distributional metrics and segmentation performance. We apply this approach to standard datasets in synthetic-to-real domain adaptation: from the video game GTA5 to real image datasets Cityscapes and ACDC, focusing on adverse conditions. Results demonstrate that evolutionary algorithms can propose diverse augmentation pipelines adapted to different objectives. The contribution of this work is the formulation of style transfer as a sequencing problem suitable for evolutionary optimization and the study of efficient metrics that enable feasible search in this space. The source code is available at: https://github.com/echigot/MOOSS.

MLV-Edit: Towards Consistent and Highly Efficient Editing for Minute-Level Videos

We propose MLV-Edit, a training-free, flow-based framework that address the unique challenges of minute-level video editing. While existing techniques excel in short-form video manipulation, scaling them to long-duration videos remains challenging due to prohibitive computational overhead and the difficulty of maintaining global temporal consistency across thousands of frames. To address this, MLV-Edit employs a divide-and-conquer strategy for segment-wise editing, facilitated by two core modules: Velocity Blend rectifies motion inconsistencies at segment boundaries by aligning the flow fields of adjacent chunks, eliminating flickering and boundary artifacts commonly observed in fragmented video processing; and Attention Sink anchors local segment features to global reference frames, effectively suppressing cumulative structural drift. Extensive quantitative and qualitative experiments demonstrate that MLV-Edit consistently outperforms state-of-the-art methods in terms of temporal stability and semantic fidelity.

Sem-NaVAE: Semantically-Guided Outdoor Mapless Navigation via Generative Trajectory Priors

This work presents a mapless global navigation approach for outdoor applications. It combines the exploratory capacity of conditional variational autoencoders (CVAEs) to generate trajectories and the semantic segmentation capabilities of a lightweight visual language model (VLM) to select the trajectory to execute. Open-vocabulary segmentation is used to score and select the generated trajectories based on natural language, and a state-of-the-art local planner executes velocity commands. One of the key features of the proposed approach is its ability to generate a large variability of trajectories and to select them and navigate in real-time. The approach was validated through real-world outdoor navigation experiments, achieving superior performance compared to state-of-the-art methods. A video showing an experimental run of the system can be found in https://www.youtube.com/watch?v=i3R5ey5O2yk.

InspecSafe-V1: A Multimodal Benchmark for Safety Assessment in Industrial Inspection Scenarios

With the rapid development of industrial intelligence and unmanned inspection, reliable perception and safety assessment for AI systems in complex and dynamic industrial sites has become a key bottleneck for deploying predictive maintenance and autonomous inspection. Most public datasets remain limited by simulated data sources, single-modality sensing, or the absence of fine-grained object-level annotations, which prevents robust scene understanding and multimodal safety reasoning for industrial foundation models. To address these limitations, InspecSafe-V1 is released as the first multimodal benchmark dataset for industrial inspection safety assessment that is collected from routine operations of real inspection robots in real-world environments. InspecSafe-V1 covers five representative industrial scenarios, including tunnels, power facilities, sintering equipment, oil and gas petrochemical plants, and coal conveyor trestles. The dataset is constructed from 41 wheeled and rail-mounted inspection robots operating at 2,239 valid inspection sites, yielding 5,013 inspection instances. For each instance, pixel-level segmentation annotations are provided for key objects in visible-spectrum images. In addition, a semantic scene description and a corresponding safety level label are provided according to practical inspection tasks. Seven synchronized sensing modalities are further included, including infrared video, audio, depth point clouds, radar point clouds, gas measurements, temperature, and humidity, to support multimodal anomaly recognition, cross-modal fusion, and comprehensive safety assessment in industrial environments.

LEMON: How Well Do MLLMs Perform Temporal Multimodal Understanding on Instructional Videos?

Recent multimodal large language models (MLLMs) have shown remarkable progress across vision, audio, and language tasks, yet their performance on long-form, knowledge-intensive, and temporally structured educational content remains largely unexplored. To bridge this gap, we introduce LEMON, a Lecture-based Evaluation benchmark for MultimOdal uNderstanding, focusing on STEM lecture videos that require long-horizon reasoning and cross-modal integration. LEMON comprises 2,277 video segments spanning 5 disciplines and 29 courses, with an average duration of 196.1 seconds, yielding 4,181 high-quality QA pairs, including 3,413 multiple-choice and 768 open-ended questions. Distinct from existing video benchmarks, LEMON features: (1) semantic richness and disciplinary density, (2) tightly coupled video-audio-text modalities, (3) explicit temporal and pedagogical structure, and (4) contextually linked multi-turn questioning. It further encompasses six major tasks and twelve subtasks, covering the full cognitive spectrum from perception to reasoning and then to generation. Comprehensive experiments reveal substantial performance gaps across tasks, highlighting that even state-of-the-art MLLMs like GPT-4o struggle with temporal reasoning and instructional prediction. We expect LEMON to serve as an extensible and challenging benchmark for advancing multimodal perception, reasoning, and generation in long-form instructional contents.

PyraTok: Language-Aligned Pyramidal Tokenizer for Video Understanding and Generation

Discrete video VAEs underpin modern text-to-video generation and video understanding systems, yet existing tokenizers typically learn visual codebooks at a single scale with limited vocabularies and shallow language supervision, leading to poor cross-modal alignment and zero-shot transfer. We introduce PyraTok, a language-aligned pyramidal tokenizer that learns semantically structured discrete latents across multiple spatiotemporal resolutions. PyraTok builds on a pretrained video VAE and a novel Language aligned Pyramidal Quantization (LaPQ) module that discretizes encoder features at several depths using a shared large binary codebook, yielding compact yet expressive video token sequences. To tightly couple visual tokens with language, PyraTok jointly optimizes multi-scale text-guided quantization and a global autoregressive objective over the token hierarchy. Across ten benchmarks, PyraTok delivers state-of-the-art (SOTA) video reconstruction, consistently improves text-to-video quality, and sets new SOTA zero-shot performance on video segmentation, temporal action localization, and video understanding, scaling robustly to up to 4K/8K resolutions.

Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic Search

Long video understanding presents significant challenges for vision-language models due to extremely long context windows. Existing solutions relying on naive chunking strategies with retrieval-augmented generation, typically suffer from information fragmentation and a loss of global coherence. We present HAVEN, a unified framework for long-video understanding that enables coherent and comprehensive reasoning by integrating audiovisual entity cohesion and hierarchical video indexing with agentic search. First, we preserve semantic consistency by integrating entity-level representations across visual and auditory streams, while organizing content into a structured hierarchy spanning global summary, scene, segment, and entity levels. Then we employ an agentic search mechanism to enable dynamic retrieval and reasoning across these layers, facilitating coherent narrative reconstruction and fine-grained entity tracking. Extensive experiments demonstrate that our method achieves good temporal coherence, entity consistency, and retrieval efficiency, establishing a new state-of-the-art with an overall accuracy of 84.1% on LVBench. Notably, it achieves outstanding performance in the challenging reasoning category, reaching 80.1%. These results highlight the effectiveness of structured, multimodal reasoning for comprehensive and context-consistent understanding of long-form videos.

See More, Store Less: Memory-Efficient Resolution for Video Moment Retrieval

Recent advances in Multimodal Large Language Models (MLLMs) have improved image recognition and reasoning, but video-related tasks remain challenging due to memory constraints from dense frame processing. Existing Video Moment Retrieval (VMR) methodologies rely on sparse frame sampling, risking potential information loss, especially in lengthy videos. We propose SMORE (See MORE, store less), a framework that enhances memory efficiency while maintaining high information resolution. SMORE (1) uses query-guided captions to encode semantics aligned with user intent, (2) applies query-aware importance modulation to highlight relevant segments, and (3) adaptively compresses frames to preserve key content while reducing redundancy. This enables efficient video understanding without exceeding memory budgets. Experimental validation reveals that SMORE achieves state-of-the-art performance on QVHighlights, Charades-STA, and ActivityNet-Captions benchmarks.

CroBIM-V: Memory-Quality Controlled Remote Sensing Referring Video Object Segmentation

Remote sensing video referring object segmentation (RS-RVOS) is challenged by weak target saliency and severe visual information truncation in dynamic scenes, making it extremely difficult to maintain discriminative target representations during segmentation. Moreover, progress in this field is hindered by the absence of large-scale dedicated benchmarks, while existing models are often affected by biased initial memory construction that impairs accurate instance localization in complex scenarios, as well as indiscriminate memory accumulation that encodes noise from occlusions or misclassifications, leading to persistent error propagation. This paper advances RS-RVOS research through dual contributions in data and methodology. First, we construct RS-RVOS Bench, the first large-scale benchmark comprising 111 video sequences, about 25,000 frames, and 213,000 temporal referring annotations. Unlike common RVOS benchmarks where many expressions are written with access to the full video context, our dataset adopts a strict causality-aware annotation strategy in which linguistic references are generated solely from the target state in the initial frame. Second, we propose a memory-quality-aware online referring segmentation framework, termed Memory Quality Control with Segment Anything Model (MQC-SAM). MQC-SAM introduces a temporal motion consistency module for initial memory calibration, leveraging short-term motion trajectory priors to correct structural deviations and establish accurate memory anchoring. Furthermore, it incorporates a decoupled attention-based memory integration mechanism with dynamic quality assessment, selectively updating high-confidence semantic features while filtering unreliable information, thereby effectively preventing error accumulation and propagation. Extensive experiments on RS-RVOS Bench demonstrate that MQC-SAM achieves state-of-the-art performance.