Related papers: Token-Level Inference-Time Alignment for Vision-La…

Large vision-language models (LVLMs) remain vulnerable to hallucination, often generating content misaligned with visual inputs. Although recent training-based approaches aim to mitigate hallucination, they typically rely on predefined or…

Continuous latent-space reasoning offers a compact alternative to textual chain-of-thought for multimodal models, enabling high-dimensional visual evidence to be integrated without explicit reasoning tokens. However, we identify a…

Training a task-specific small reasoning model is challenging when direct human supervision or high-quality labels are scarce. However, LLMs with reasoning capabilities produce abundant intermediate reasoning traces that can be…

In standard large vision-language models (LVLMs) pre-training, the model typically maximizes the joint probability of the caption conditioned on the image via next-token prediction (NTP); however, since only a small subset of caption tokens…

Although Large Vision-Language Models (LVLMs) have made substantial progress, hallucination, where generated text is not grounded in the visual input, remains a challenge. As LVLMs become stronger, previously reported hallucination…

Vision-language models (VLMs) have achieved remarkable advancements, capitalizing on the impressive capabilities of large language models (LLMs) across diverse tasks. Despite this, a critical challenge known as hallucination occurs when…

Modern multimodal large language models (MLLMs) adopt a unified self-attention design that processes visual and textual tokens at every Transformer layer, incurring substantial computational overhead. In this work, we revisit the necessity…

Real-world deployment of Vision-Language Models (VLMs) is hindered by high computational demands, as existing architectures inefficiently process all tokens uniformly. We introduce Adaptive Token Pruning (ATP), a dynamic inference mechanism…

Test-time scaling (TTS) methods have proven highly effective for LLMs, yet their application to vision-language models (VLMs) remains relatively underexplored. Existing VLM TTS methods largely require open-weight model access or expensive…

Recently, large models have shown significant potential for smart healthcare. However, the deployment of Large Vision-Language Models (LVLMs) for clinical services is currently hindered by three critical challenges: a tendency to…

Vision-language models (VLMs) have demonstrated remarkable potential in integrating visual and linguistic information, but their performance is often constrained by the need for extensive, high-quality image-text training data. Curation of…

Chain-of-thought (CoT) reasoning is critical for improving the interpretability and reliability of Large Vision-Language Models (LVLMs). However, existing training algorithms such as SFT, PPO, and GRPO may not generalize well across unseen…

Multimodal language models (MLMs) still face challenges in fundamental visual perception tasks where specialized models excel. Tasks requiring reasoning about 3D structures benefit from depth estimation, and reasoning about 2D object…

Inference-time alignment provides an efficient alternative for aligning LLMs with humans. However, these approaches still face challenges, such as limited scalability due to policy-specific value functions and latency during the inference…

Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning. We study why VLMs, such as LLaVA, underutilize spatial cues despite having positional encodings and spatially rich vision encoder…

Vision-Language Models (VLM) can support clinicians by analyzing medical images and engaging in natural language interactions to assist in diagnostic and treatment tasks. However, VLMs often exhibit "hallucinogenic" behavior, generating…

Using effective generalization capabilities of vision language models (VLMs) in context-specific dynamic tasks for embodied artificial intelligence remains a significant challenge. Although supervised fine-tuned models can better align with…

Multimodal large language models (MLLMs) are prone to hallucinations, generating plausible but visually ungrounded outputs, partly because direct preference optimization (DPO) overfits to superficial linguistic cues under static preference…

Vision-Language Models (VLMs), such as CLIP, have achieved significant zero-shot performance on downstream tasks with various fine-tuning adaptation methods. However, recent studies have proven that adversarial attacks can significantly…

Vision-language models (VLMs) have recently shown promise in general-purpose reasoning tasks, yet their applicability to domain-specific scientific workflows remains largely unexplored. In this work, we evaluated a series of open-weight and…