The existing video understanding benchmarks for multimodal large language models (MLLMs) mainly focus on short videos. The few benchmarks for long video understanding often rely on multiple-choice questions (MCQs). Due to the limitations of MCQ evaluations and the advanced reasoning abilities of MLLMs, models can often answer correctly by combining short video insights with elimination, without truly understanding the content. To bridge this gap, we introduce CG-Bench, a benchmark for clue-grounded question answering in long videos. CG-Bench emphasizes the model's ability to retrieve relevant clues, enhancing evaluation credibility. It includes 1,219 manually curated videos organized into 14 primary, 171 secondary, and 638 tertiary categories, making it the largest benchmark for long video analysis. The dataset features 12,129 QA pairs in three question types: perception, reasoning, and hallucination. To address the limitations of MCQ-based evaluation, we develop two novel clue-based methods: clue-grounded white box and black box evaluations, assessing whether models generate answers based on accurate video understanding. We evaluated multiple closed-source and open-source MLLMs on CG-Bench. The results show that current models struggle significantly with long videos compared to short ones, and there is a notable gap between open-source and commercial models. We hope CG-Bench will drive the development of more reliable and capable MLLMs for long video comprehension.

In egocentric video understanding, the motion of hands and objects as well as their interactions play a significant role by nature. However, existing egocentric video representation learning methods mainly focus on aligning video representation with high-level narrations, overlooking the intricate dynamics between hands and objects. In this work, we aim to integrate the modeling of fine-grained hand-object dynamics into the video representation learning process. Since no suitable data is available, we introduce HOD, a novel pipeline employing a hand-object detector and a large language model to generate high-quality narrations with detailed descriptions of hand-object dynamics. To learn these fine-grained dynamics, we propose EgoVideo, a model with a new lightweight motion adapter to capture fine-grained hand-object motion information. Through our co-training strategy, EgoVideo effectively and efficiently leverages the fine-grained hand-object dynamics in the HOD data. Extensive experiments demonstrate that our method achieves state-of-the-art performance across multiple egocentric downstream tasks, including improvements of 6.3% in EK-100 multi-instance retrieval, 5.7% in EK-100 classification, and 16.3% in EGTEA classification in zero-shot settings. Furthermore, our model exhibits robust generalization capabilities in hand-object interaction and robot manipulation tasks.

Transferring and integrating knowledge across first-person (egocentric) and third-person (exocentric) viewpoints is intrinsic to human intelligence, enabling humans to learn from others and convey insights from their own experiences. Despite rapid progress in multimodal large language models (MLLMs), their ability to perform such cross-view reasoning remains unexplored. To address this, we introduce EgoExoBench, the first benchmark for egocentric exocentric video understanding and reasoning. Built from publicly available datasets, EgoExoBench comprises over 7300 question–answer pairs spanning eleven sub-tasks organized into three core challenges: semantic alignment, viewpoint association, and temporal reasoning. We evaluate 13 state-of-the-art MLLMs and find that while these models excel on single-view tasks, they struggle to align semantics across perspectives, accurately associate views, and infer temporal dynamics in the ego-exo context. We hope EgoExoBench can serve as a valuable resource for research on embodied agents and intelligent assistants seeking human-like cross-view intelligence.

Egocentric video reasoning centers on an unobservable agent behind the camera who dynamically shapes the environment, requiring inference of hidden intentions and recognition of fine-grained interactions. This core challenge limits current multimodal large language models (MLLMs), which excel at visible event reasoning but lack embodied, first-person understanding. To bridge this gap, we introduce EgoThinker, a novel framework that endows MLLMs with robust egocentric reasoning capabilities through spatio-temporal chain-of-thought supervision and a two-stage learning curriculum. First, we introduce EgoRe-5M, a large-scale egocentric QA dataset constructed from 13M diverse egocentric video clips. This dataset features multi-minute segments annotated with detailed CoT rationales and dense hand–object grounding. Second, we employ SFT on EgoRe-5M to instill reasoning skills, followed by reinforcement fine-tuning (RFT) to further enhance spatio-temporal localization. Experimental results show that EgoThinker outperforms existing methods across multiple egocentric benchmarks, while achieving substantial improvements in fine-grained spatio-temporal localization tasks.

In the fields of computer vision and robotics, accurate pixel-level correspondences are essential for enabling advanced tasks such as structure-from-motion and simultaneous localization and mapping. Recent correspondence pruning methods usually focus on learning local consistency through k-nearest neighbors, which makes it difficult to capture robust context for each correspondence. We propose CorrAdaptor, a novel architecture that introduces a dual-branch structure capable of adaptively adjusting local contexts through both explicit and implicit local graph learning. Specifically, the explicit branch uses KNN-based graphs tailored for initial neighborhood identification, while the implicit branch leverages a learnable matrix to softly assign neighbors and adaptively expand the local context scope, significantly enhancing the model’s robustness and adaptability to complex image variations. Moreover, we design a motion injection module to integrate motion consistency into the network to suppress the impact of outliers and refine local context learning, resulting in substantial performance improvements. The experimental results on extensive correspondence-based tasks indicate that our CorrAdaptor achieves state-of-the-art performance both qualitatively and quantitatively.

This paper presents a unified lateral preview driver model for closed-loop dynamic simulations of road vehicles. Numerous driver models have been proposed for Single-Unit Vehicles (SUVs). Some SUV-based driver models have been applied to closed-loop simulations of Multi-Trailer Articulated Heavy Vehicles (MTAHVs). However, the dynamics of MTAHVs is significantly different from that of SUVs, and drivers of Multi-Unit Vehicles (MUVs) have different driving performance and skills. Very few driver models have been proposed for closed-loop simulations of MUVs. This paper designs the unified driver model, considering the dynamic features of both SUVs and MUVs. The driver model is derived using a sliding mode control (SMC) technique, and it distinguishes itself from conventional driver models with a number of features. Simulations demonstrate the applicability and effectiveness of the proposed driver model.

Control algorithms are essential in the modern world to tackle with ever-changing environmental disturbances while meeting adequate safety standards. Control systems must be appropriately tuned for each application to ensure reliability and safety. This paper applies a multi-objective optimization method, Generalized Differential Evolution (GDE3), to tune a Linear Quadratic Regulator (LQR) while enabling posteriori decision-making. The utilized case study is the aircraft pitch control. We compare the results of GDE3 with Non-Dominated Sorting Genetic Algorithm (NSGA-II) and conventional tuning. The current findings show that the GDE3 performs better than the other methods. This paper sheds light on how trade-off and condition-specific based optimization can enhance real-time control systems.

In this paper, we study the effect of the dedicated short range communication (DSRC) system on the stability of an articulated heavy vehicle that is using active trailer steering to improve the lateral stability under a double lane change maneuverability test. Several design parameters of the communication system such as modulation, quantization, channel estimation algorithm and transmit diversity have been studied and the effect of each parameter on the active control system is illustrated through simulations. The numerical evaluations show that, by adopting proper configurations, the wireless communication system can perform as good as the wired system it replaces in terms of successfully stabilizing the vehicle.

A charge-based on-chip synapse adaptation Kohonen neural network circuit is proposed. The properties of the approach are low power dissipation and high density due to the charge transfer mechanism and the novel compact device configurations. The prototype chip which contains 12x10 synapses with a density of 190 synapses/mm(2 ) was fabricated with 2-mum standard CMOS technology. The experimental results from the prototype chip demonstrated successful unsupervised learning and classification as theoretically predicted.

A charge-based programmable Hamming neural network circuit is proposed. It utilizes capacitive comparators as processing elements in the feedforward layer, and a multiport charge-sensing amplifier as the MAXNET (or winner-take-all (WTA)) circuit. The CMOS prototype chip contains 10*10 fully interconnected processing elements with the capability of encoding 10 exemplar patterns. The whole circuit occupies a silicon area of 0.414 mm/sup 2/ fabricated in a 2- mu m CMOS technology. The low-silicon area and low-power dissipation are the fundamental properties of the proposed implementation. The experimental results from a prototype chip show robust retrieval and excellent classification properties as theoretically predicted. A modularity methodology and how to extend the prototype chip to VLSI system level integration are examined.>