Related papers: Entropy-Gradient Grounding: Training-Free Evidence…

Fine-tuning MLLMs for Video Temporal Grounding (VTG) often improves in-domain performance but degrades sharply under domain shift. In this work, we find that this failure is primarily driven not just by unseen query concepts, but by visual…

Video temporal grounding aims to identify video segments within untrimmed videos that are most relevant to a given natural language query. Existing video temporal localization models rely on specific datasets for training and have high data…

Vision-Language Models (VLMs) learn joint representations by mapping images and text into a shared latent space. However, recent research highlights that deterministic embeddings from standard VLMs often struggle to capture the…

Vision-Language Models (VLMs) can generate convincing clinical narratives, yet frequently struggle to visually ground their statements. We posit this limitation arises from the scarcity of high-quality, large-scale clinical…

Large vision language models (VLMs) increasingly claim reasoning skills, yet current benchmarks evaluate them in single-turn or question answering settings. However, grounding is an interactive process in which people gradually develop…

Vision-Language Models (VLMs) demonstrate impressive capabilities across multimodal tasks, yet exhibit systematic spatial reasoning failures, achieving only 49% (CLIP) to 54% (BLIP-2) accuracy on basic directional relationships. For safe…

Models like OpenAI-o3 pioneer visual grounded reasoning by dynamically referencing visual regions, just like human "thinking with images". However, no benchmark exists to evaluate these capabilities holistically. To bridge this gap, we…

Grounding language in vision is an active field of research seeking to construct cognitively plausible word and sentence representations by incorporating perceptual knowledge from vision into text-based representations. Despite many…

Visual grounding, localizing objects from natural language descriptions, represents a critical bridge between language and vision understanding. While multimodal large language models (MLLMs) achieve impressive scores on existing…

Multi-step processes via large language models (LLMs) have proven effective for solving complex reasoning tasks. However, the depth of exploration of the reasoning procedure can significantly affect the task performance. Existing methods to…

Recent multimodal large language models (MLLMs) have begun to support Thinking with Images by invoking visual tools such as zooming and cropping during inference. Yet these systems remain brittle in fine-grained visual reasoning because…

Temporal Video Grounding (TVG), the task of locating specific video segments based on language queries, is a core challenge in long-form video understanding. While recent Large Vision-Language Models (LVLMs) have shown early promise in…

Large Vision-Language Models (LVLMs) have become powerful general-purpose assistants, yet their predictions often lack reliability and interpretability due to insufficient grounding in visual evidence. The emerging thinking-with-images…

Joint video-language learning has received increasing attention in recent years. However, existing works mainly focus on single or multiple trimmed video clips (events), which makes human-annotated event boundaries necessary during…

Visual-language models (VLMs) have recently been introduced in robotic mapping using the latent representations, i.e., embeddings, of the VLMs to represent semantics in the map. They allow moving from a limited set of human-created labels…

Test-time adaptation paradigm provides flexibility towards domain shifts by performing immediate adaptation on unlabeled target data from the source model. Vision-Language Models (VLMs) leverage their generalization capabilities for diverse…

We introduce a language-grounded visual prompting method to adapt the visual encoder of vision-language models for downstream tasks. By capitalizing on language integration, we devise a parameter-efficient strategy to adjust the input of…

In this paper, the LCV2 modular method is proposed for the Grounded Visual Question Answering task in the vision-language multimodal domain. This approach relies on a frozen large language model (LLM) as intermediate mediator between the…

We introduce Entropy2Vec, a novel framework for deriving cross-lingual language representations by leveraging the entropy of monolingual language models. Unlike traditional typological inventories that suffer from feature sparsity and…

We study the feasibility of identifying epistemic uncertainty (reflecting a lack of knowledge), as opposed to aleatoric uncertainty (reflecting entropy in the underlying distribution), in the outputs of large language models (LLMs) over…