Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs

• URL: http://arxiv.org/abs/2510.00705v1
• Date: Wed, 01 Oct 2025 09:20:51 GMT
• Title: Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs
• Authors: Sanghwan Kim, Rui Xiao, Stephan Alaniz, Yongqin Xian, Zeynep Akata,
• Abstract summary: We propose a training-free framework that uses an MLLM's intrinsic uncertainty as a proactive guidance signal.<n>We introduce a unified mechanism that scores candidate visual inputs by response uncertainty, enabling the model to autonomously focus on the most salient data.<n>Our work validates that harnessing intrinsic uncertainty is a powerful, general strategy for enhancing fine-grained multimodal performance.
• Score: 61.64185573373394
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal Large Language Models (MLLMs) often struggle with fine-grained perception, such as identifying small objects in high-resolution images or finding key moments in long videos. Existing works typically rely on complicated, task-specific fine-tuning, which limits their generalizability and increases model complexity. In this work, we propose an effective, training-free framework that uses an MLLM's intrinsic uncertainty as a proactive guidance signal. Our core insight is that a model's output entropy decreases when presented with relevant visual information. We introduce a unified mechanism that scores candidate visual inputs by response uncertainty, enabling the model to autonomously focus on the most salient data. We apply this simple principle to three complex visual tasks: Visual Search, Long Video Understanding, and Temporal Grounding, allowing off-the-shelf MLLMs to achieve performance competitive with specialized, fine-tuned methods. Our work validates that harnessing intrinsic uncertainty is a powerful, general strategy for enhancing fine-grained multimodal performance.

Related papers

• MVR: Multi-view Video Reward Shaping for Reinforcement Learning [17.20077949643041]
  Multi-View Video Reward Shaping (MVR) is a framework that models the relevance of states regarding the target task using videos captured from multiple viewpoints.<n>MVR learns a state relevance function that mitigates the bias towards specific static poses inherent in image-based methods.<n>We introduce a state-dependent reward shaping formulation that integrates task-specific rewards and VLM-based guidance.
  arXiv  Detail & Related papers  (2026-03-02T10:24:04Z)
• Magic-MM-Embedding: Towards Visual-Token-Efficient Universal Multimodal Embedding with MLLMs [10.443777669301983]
  Multimodal Large Language Models (MLLMs) have shown immense promise in universal multimodal retrieval.<n>But their practical application is often hindered by the substantial computational cost incurred from processing a large number of tokens from visual inputs.<n>We propose Magic-MM-Embedding, a series of novel models that achieve both high efficiency and state-of-the-art performance in universal multimodal embedding.
  arXiv  Detail & Related papers  (2026-02-05T04:01:01Z)
• A Comprehensive Study on Visual Token Redundancy for Discrete Diffusion-based Multimodal Large Language Models [85.30893355216486]
  We study how visual token redundancy evolves with different dMLLM architectures and tasks.<n>Our study reveals that visual redundancy emerges only in from-scratch dMLLMs while handling long-answer tasks.<n>Layer-skipping is promising for accelerating AR-to-diffusion dMLLMs, whereas progressive or late-step pruning is more effective for from-scratch dMLLMs.
  arXiv  Detail & Related papers  (2025-11-19T04:13:36Z)
• Visual Representation Alignment for Multimodal Large Language Models [38.319869213758686]
  Multimodal large language models (MLLMs) trained with visual instruction tuning have achieved strong performance across diverse tasks.<n>But they remain limited in vision-centric tasks such as object counting or spatial reasoning.<n>We present VIsual Representation ALignment (VIRAL), a simple yet effective regularization strategy that aligns the internal visual representations of MLLMs with those of pre-trained vision foundation models.
  arXiv  Detail & Related papers  (2025-09-09T17:59:14Z)
• LFTR: Learning-Free Token Reduction for Multimodal Large Language Models [3.368594680297987]
  We introduce a learning-free token reduction (LFTR) method designed for Multimodal Large Language Models (MLLMs)<n>By capitalizing on the redundancy in visual representations, our approach effectively reduces tokens while preserving the general inference performance of MLLMs.<n>Our results show that LFTR achieves up to a $16times$ reduction of visual tokens while maintaining or even enhancing performance on mainstream vision question-answering benchmarks.
  arXiv  Detail & Related papers  (2025-01-29T02:52:32Z)
• FOLDER: Accelerating Multi-modal Large Language Models with Enhanced Performance [9.782362715017596]
  We introduce FOLDER, a simple yet effective plug-and-play module designed to reduce the length of the visual token sequence.<n>We analyze the information loss introduced by different reduction strategies and develop FOLDER to preserve key information while removing visual redundancy.<n>FOLDER achieves comparable or even better performance than the original models, while dramatically reducing complexity by removing up to 70% of visual tokens.
  arXiv  Detail & Related papers  (2025-01-05T03:28:45Z)
• Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment [58.94611347128066]
  multimodal large language models (MLLMs) struggle with fine-grained or precise understanding of visuals.<n>Recent studies either develop tool-using or unify specific visual tasks into the autoregressive framework, often at the expense of overall multimodal performance.<n>We propose Task Preference Optimization (TPO), a novel method that utilizes differentiable task preferences derived from typical fine-grained visual tasks.
  arXiv  Detail & Related papers  (2024-12-26T18:56:05Z)
• TWIST & SCOUT: Grounding Multimodal LLM-Experts by Forget-Free Tuning [54.033346088090674]
  We introduce TWIST & SCOUT, a framework that equips pre-trained MLLMs with visual grounding ability.<n>To fine-tune the model effectively, we generate a high-quality synthetic dataset we call SCOUT.<n>This dataset provides rich supervision signals, describing a step-by-step multimodal reasoning process.
  arXiv  Detail & Related papers  (2024-10-14T13:35:47Z)
• Multi-granularity Contrastive Cross-modal Collaborative Generation for End-to-End Long-term Video Question Answering [53.39158264785098]
  Long-term Video Question Answering (VideoQA) is a challenging vision-and-language bridging task. We present an entirely end-to-end solution for VideoQA: Multi-granularity Contrastive cross-modal collaborative Generation model.
  arXiv  Detail & Related papers  (2024-10-12T06:21:58Z)
• NoteLLM-2: Multimodal Large Representation Models for Recommendation [71.87790090964734]
  Large Language Models (LLMs) have demonstrated exceptional proficiency in text understanding and embedding tasks.<n>Their potential in multimodal representation, particularly for item-to-item (I2I) recommendations, remains underexplored.<n>We propose an end-to-end fine-tuning method that customizes the integration of any existing LLMs and vision encoders for efficient multimodal representation.
  arXiv  Detail & Related papers  (2024-05-27T03:24:01Z)
• DoraemonGPT: Toward Understanding Dynamic Scenes with Large Language Models (Exemplified as A Video Agent) [73.10899129264375]
  This paper explores DoraemonGPT, a comprehensive and conceptually elegant system driven by LLMs to understand dynamic scenes.<n>Given a video with a question/task, DoraemonGPT begins by converting the input video into a symbolic memory that stores task-related attributes.<n>We extensively evaluate DoraemonGPT's effectiveness on three benchmarks and several in-the-wild scenarios.
  arXiv  Detail & Related papers  (2024-01-16T14:33:09Z)
• PALM: Predicting Actions through Language Models [74.10147822693791]
  We introduce PALM, an approach that tackles the task of long-term action anticipation. Our method incorporates an action recognition model to track previous action sequences and a vision-language model to articulate relevant environmental details. Our experimental results demonstrate that PALM surpasses the state-of-the-art methods in the task of long-term action anticipation.
  arXiv  Detail & Related papers  (2023-11-29T02:17:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.