The paper introduces EASE, a method that enhances multimodal Reinforcement Learning with Verifiable Rewards (RLVR) by providing spatial attention supervision anchored to visual evidence, significantly…

The paper addresses Perceptual Judgment Bias in multimodal LLM judges by introducing a new dataset and a unified training framework that forces models to prioritize visual evidence over plausible text…

The paper proposes a training-free framework, Visual Representation-Guided Video-LLM Reasoning, to perform composed video retrieval by using visual examples and text instructions, achieving strong per…

The paper introduces Responsible Contrastive Soft Prompting (RCSP), a parameter-efficient method using soft prompts to improve LLM reliability by simultaneously suppressing hallucinations, encouraging…

The paper deconstructs latent visual reasoning tokens into components and finds that the performance gains are primarily due to boundary markers and attention patterns, not the tokens' ability to enco…

The paper introduces StemBind, a diagnostic benchmark that separates perception, rule induction, and answer selection in abstract visual reasoning, revealing that the primary failure point for MLLMs i…

The paper proposes EAGLE, a novel evidence-aligned multi-agent framework, demonstrating that requiring shared visual evidence among agents is crucial for achieving reliable and trustworthy consensus i…

The paper proposes a unified framework that decouples long-video reasoning into semantic and visual evidence, significantly improving performance on the HD-EPIC VQA Challenge.

OccamToken introduces a training-free, adaptive token pruning framework that replaces fixed token budgets with relative evidence testing against a register-based reference, significantly improving VLM…

The paper introduces a robust, two-part framework (HyPE and HyPS) using hyperbolic geometry to efficiently detect and sanitize malicious prompts targeting Vision-Language Models (VLMs).

The paper proposes CSMR, a cognitive scheduling framework that allows a language model to dynamically decide when to acquire task-relevant visual evidence, significantly improving multimodal reasoning…

VITAL introduces a novel latent-space reasoning framework for medical MLLMs, utilizing visual-semantic dual supervision to enhance reasoning capabilities and provide crucial interpretability without s…

The paper systematically evaluates concept-based explainability in MLLMs, finding that forcing models to generate formal explanations degrades predictive accuracy, suggesting that explaining is genuin…

The paper introduces MLLM-Microscope, a system that analyzes the internal structure of multimodal large language models (MLLMs), finding that modality fusion significantly impacts the linearity and di…

This paper introduces a new evaluation framework, SpatialUncertain, demonstrating that current Vision-Language Models (VLMs) are prone to overconfident and incorrect answers to spatial questions when…

PARCEL introduces a novel visual tokenization architecture that combines spatial pooling anchors with conditioned elastic queries, efficiently reducing the computational cost of large Vision-Language…

The paper proposes Visual Gradient Steering (VGS), a method that decomposes the distillation loss into language and visual components and steers the optimization to prioritize visual grounding, signif…

The paper proposes using Vision-Language Models (VLMs) as 'teachers' to guide Video Generation Models (VGMs) during test-time optimization, significantly improving video reasoning capabilities.

The paper introduces TRON, an online, rule-verifiable environment substrate that generates an unbounded stream of fresh, controllable visual reasoning training instances, significantly improving RL pe…