This work explores enabling Chain-of-Thought (CoT) reasoning to link visual cues across multiple images. A straightforward solution is to adapt rule-based reinforcement learning for Vision-Language Models (VLMs). However, such methods typically rely on manually curated question-answer pairs, which can be particularly challenging when dealing with fine-grained visual details and complex logic across images. Inspired by self-supervised visual representation learning, we observe that images contain inherent constraints that can serve as supervision. Based on this insight, we construct image triplets comprising two augmented views of the same image and a third, similar but distinct image. During training, the model is prompted to generate a reasoning process to compare these images (i.e., determine same or different).

Egocentric video-language pretraining has significantly advanced video representation learning. Humans perceive and interact with a fully 3D world, developing spatial awareness that extends beyond text-based understanding. However, most previous works learn from 1D text or 2D visual cues, such as bounding boxes, which inherently lack 3D understanding. To bridge this gap, we introduce EgoDTM, an Egocentric Depth-and Text-aware Model, jointly trained through large-scale 3D-aware video pretraining and video-text contrastive learning. EgoDTM incorporates a lightweight 3D-aware decoder to efficiently learn 3D-awareness from pseudo depth maps generated by depth estimation models. To further facilitate 3D-aware video pretraining, we enrich the original brief captions with hand-object visual cues by organically combining several foundation models. Extensive experiments demonstrate EgoDTM's superior performance across diverse downstream tasks, highlighting its superior 3D-aware visual understanding.

To stop children from bypassing its age checks, Meta is revamping its age-verification tools with an AI system that analyzes images and videos for "visual cues," such as height and bone structure. Meta is beefing up its age-verification mechanisms with an AI system that analyzes images and videos on Instagram and Facebook for "visual cues," such as height and bone structure, to identify and delete accounts of users under the age of 13. The company announced the move amid a wave of cases in which hundreds of children have managed to evade social network access restrictions, even through simple tricks such as drawing on a mustache. The new approach is part of a series of measures Meta adopted as part of an AI-based security strategy designed to correct the limitations of traditional methods, which rely heavily on self-reported age. With this change, the company seeks to reduce the ease with which minors access platforms that, in theory, are restricted to them.

Audio-visual target speaker extraction (AV-TSE) models primarily rely on visual cues from the target speaker. However, humans also leverage linguistic knowledge, such as syntactic constraints, next word prediction, and prior knowledge of conversation, to extract target speech. Inspired by this observation, we propose ELEGANCE, a novel framework that incorporates linguistic knowledge from large language models (LLMs) into AV-TSE models through three distinct guidance strategies: output linguistic constraints, intermediate linguistic prediction, and input linguistic prior. Comprehensive experiments with RoBERTa, Qwen3-0.6B, and Qwen3-4B on two AV-TSE backbones demonstrate the effectiveness of our approach. Significant improvements are observed in challenging scenarios, including visual cue impaired, unseen languages, target speaker switches, increased interfering speakers, and out-of-domain test set. Demo page: https://alexwxwu.github.io/ELEGANCE/.

Accurate predictive turn-taking models (PTTMs) are essential for naturalistic human-robot interaction. However, little is known about their performance in noise. This study therefore explores PTTM performance in types of noise likely to be encountered once deployed. Our analyses reveal PTTMs are highly sensitive to noise. Hold/shift accuracy drops from 84% in clean speech to just 52% in 10 dB music noise. Training with noisy data enables a multimodal PTTM, which includes visual features to better exploit visual cues, with 72% accuracy in 10 dB music noise. The multimodal PTTM outperforms the audio-only PTTM across all noise types and SNRs, highlighting its ability to exploit visual cues; however, this does not always generalise to new types of noise. Analysis also reveals that successful training relies on accurate transcription, limiting the use of ASR-derived transcriptions to clean conditions. We make code publicly available for future research.

Turn-taking is richly multimodal. Predictive turn-taking models (PTTMs) facilitate naturalistic human-robot interaction, yet most rely solely on speech. We introduce MM-VAP, a multimodal PTTM which combines speech with visual cues including facial expression, head pose and gaze. We find that it outperforms the state-of-the-art audio-only in videoconferencing interactions (84% vs. 79% hold/shift prediction accuracy). Unlike prior work which aggregates all holds and shifts, we group by duration of silence between turns. This reveals that through the inclusion of visual features, MM-VAP outperforms a state-of-the-art audio-only turn-taking model across all durations of speaker transitions. We conduct a detailed ablation study, which reveals that facial expression features contribute the most to model performance. Thus, our working hypothesis is that when interlocutors can see one another, visual cues are vital for turn-taking and must therefore be included for accurate turn-taking prediction. We additionally validate the suitability of automatic speech alignment for PTTM training using telephone speech. This work represents the first comprehensive analysis of multimodal PTTMs. We discuss implications for future work and make all code publicly available.

Pseudo - haptics exploit carefully crafted visual or auditory cues to trick the brain into "feeling" forces that are never physically applied, offering a low - cost alternative to traditional haptic hardware. Here, we present a comparative psychophysical study that quantifies how visual and auditory stimuli combine to evoke pseudo - haptic pressure sensations on a commodity tablet. Using a Unity - based Rollball game, participants (n = 4) guided a virtual ball across three textured terrains while their fi nger forces were captured in real time with a Robotous RFT40 force - torque sensor. Each terrain was paired with a distinct rolling - sound profile spanning 440 Hz - 4.7 kHz, 440 Hz - 13.1 kHz, or 440 Hz - 8.9 kHz; crevice collisions triggered additional "knoc king" bursts to heighten realism. Average tactile forces increased systematically with cue intensity: 0.40 N, 0.79 N and 0.88 N for visual - only trials and 0.41 N, 0.81 N and 0.90 N for audio - only trials on Terrains 1 - 3, respectively. Higher audio frequenci es and denser visual textures both elicited stronger muscle activation, and their combination further reduced the force needed to perceive surface changes, confirming multisensory integration. These results demonstrate that consumer - grade isometric devices can reliably induce, and measure graded pseudo - haptic feedback without specialized actuators, opening a path toward affordable rehabilitation tools, training simulators and assistive interfaces .

Then, it leverages large language models (LLMs) to generate description-based prompts (or visual cues) for each component.

The advent of autonomous driving systems promises to transform transportation by enhancing safety, efficiency, and comfort. As these technologies evolve toward higher levels of autonomy, the need for integrated systems that seamlessly support human involvement in decision-making becomes increasingly critical. Certain scenarios necessitate human involvement, including those where the vehicle is unable to identify an object or element in the scene, and as such cannot take independent action. Therefore, situational awareness is essential to mitigate potential risks during a takeover, where a driver must assume control and autonomy from the vehicle. The need for driver attention is important to avoid collisions with external agents and ensure a smooth transition during takeover operations. This paper explores the integration of attention redirection techniques, such as gaze manipulation through targeted visual and auditory cues, to help drivers maintain focus on emerging hazards and reduce target fixation in semi-autonomous driving scenarios. We propose a conceptual framework that combines real-time gaze tracking, context-aware saliency analysis, and synchronized visual and auditory alerts to enhance situational awareness, proactively address potential hazards, and foster effective collaboration between humans and autonomous systems.