Instruction-Oriented Preference Alignment for Enhancing Multi-Modal Comprehension Capability of MLLMs

• URL: http://arxiv.org/abs/2503.20309v1
• Date: Wed, 26 Mar 2025 08:19:02 GMT
• Title: Instruction-Oriented Preference Alignment for Enhancing Multi-Modal Comprehension Capability of MLLMs
• Authors: Zitian Wang, Yue Liao, Kang Rong, Fengyun Rao, Yibo Yang, Si Liu,
• Abstract summary: We propose a scalable framework designed to automatically construct alignment preferences grounded in instruction fulfillment efficacy.<n>Our method involves an automated preference construction coupled with a dedicated verification process.<n>Experiments conducted on Qwen2VL-7B demonstrate IPA's effectiveness across multiple benchmarks.
• Score: 29.07102440466282
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Preference alignment has emerged as an effective strategy to enhance the performance of Multimodal Large Language Models (MLLMs) following supervised fine-tuning. While existing preference alignment methods predominantly target hallucination factors, they overlook the factors essential for multi-modal comprehension capabilities, often narrowing their improvements on hallucination mitigation. To bridge this gap, we propose Instruction-oriented Preference Alignment (IPA), a scalable framework designed to automatically construct alignment preferences grounded in instruction fulfillment efficacy. Our method involves an automated preference construction coupled with a dedicated verification process that identifies instruction-oriented factors, avoiding significant variability in response representations. Additionally, IPA incorporates a progressive preference collection pipeline, further recalling challenging samples through model self-evolution and reference-guided refinement. Experiments conducted on Qwen2VL-7B demonstrate IPA's effectiveness across multiple benchmarks, including hallucination evaluation, visual question answering, and text understanding tasks, highlighting its capability to enhance general comprehension.

Related papers

• Enhancing LLM Robustness to Perturbed Instructions: An Empirical Study [8.827173113748701]
  We study character- and word-level edits of task-specific instructions, which substantially degrade downstream performance. We find that, on average, self-denoising achieves substantially higher performance gains than alternative strategies.
  arXiv  Detail & Related papers  (2025-04-03T16:17:56Z)
• Re-Align: Aligning Vision Language Models via Retrieval-Augmented Direct Preference Optimization [19.37373012848517]
  Large Vision Language Models (VLMs) are prone to significant hallucinations, particularly in the form of cross-modal inconsistencies.<n>We introduce Re-Align, a novel alignment framework that leverages image retrieval to construct a dual-preference dataset.<n>We also introduce rDPO, an extension of the standard direct preference optimization that incorporates an additional visual preference objective during fine-tuning.
  arXiv  Detail & Related papers  (2025-02-18T18:59:57Z)
• Towards more Contextual Agents: An extractor-Generator Optimization Framework [0.0]
  Large Language Model (LLM)-based agents have demonstrated remarkable success in solving complex tasks across a wide range of general-purpose applications.<n>However, their performance often degrades in context-specific scenarios, such as specialized industries or research domains.<n>To address this challenge, our work introduces a systematic approach to enhance the contextual adaptability of LLM-based agents.
  arXiv  Detail & Related papers  (2025-02-18T15:07:06Z)
• Reward-aware Preference Optimization: A Unified Mathematical Framework for Model Alignment [45.45508377432791]
  This paper introduces Reward-Aware Preference Optimization (RPO), a mathematical framework that unifies popular preference optimization techniques.<n>RPO provides a structured approach to disentangle and systematically study the impact of various design choices.<n>We propose a new experimental setup that enables the clean and direct ablation of such design choices.
  arXiv  Detail & Related papers  (2025-01-31T22:39:04Z)
• CHiP: Cross-modal Hierarchical Direct Preference Optimization for Multimodal LLMs [107.21334626890713]
  Multimodal Large Language Models (MLLMs) still struggle with hallucinations despite their impressive capabilities. We propose a Cross-modal Hierarchical Direct Preference Optimization (CHiP) to address these limitations. We evaluate CHiP through both quantitative and qualitative analyses, with results across multiple benchmarks demonstrating its effectiveness in reducing hallucinations.
  arXiv  Detail & Related papers  (2025-01-28T02:05:38Z)
• Teaching LLMs to Refine with Tools [68.23479664749271]
  Large language models (LLMs) can refine their responses based on feedback, enabling self-improvement through iterative training or test-time refinement.<n>We propose CaP, a novel approach that uses external tools to refine chain-of-thought (CoT) responses generated by the same or other LLMs.
  arXiv  Detail & Related papers  (2024-12-22T05:43:50Z)
• A Systematic Examination of Preference Learning through the Lens of Instruction-Following [83.71180850955679]
  We use a novel synthetic data generation pipeline to generate 48,000 instruction unique-following prompts.<n>With our synthetic prompts, we use two preference dataset curation methods - rejection sampling (RS) and Monte Carlo Tree Search (MCTS)<n>Experiments reveal that shared prefixes in preference pairs, as generated by MCTS, provide marginal but consistent improvements.<n>High-contrast preference pairs generally outperform low-contrast pairs; however, combining both often yields the best performance.
  arXiv  Detail & Related papers  (2024-12-18T15:38:39Z)
• Debiasing Multimodal Large Language Models [61.6896704217147]
  Large Vision-Language Models (LVLMs) have become indispensable tools in computer vision and natural language processing. Our investigation reveals a noteworthy bias in the generated content, where the output is primarily influenced by the underlying Large Language Models (LLMs) prior to the input image. To rectify these biases and redirect the model's focus toward vision information, we introduce two simple, training-free strategies.
  arXiv  Detail & Related papers  (2024-03-08T12:35:07Z)
• Relative Preference Optimization: Enhancing LLM Alignment through Contrasting Responses across Identical and Diverse Prompts [95.09994361995389]
  Relative Preference Optimization (RPO) is designed to discern between more and less preferred responses derived from both identical and related prompts. RPO has demonstrated a superior ability to align large language models with user preferences and to improve their adaptability during the training process.
  arXiv  Detail & Related papers  (2024-02-12T22:47:57Z)
• Entropy-Regularized Token-Level Policy Optimization for Language Agent Reinforcement [67.1393112206885]
  Large Language Models (LLMs) have shown promise as intelligent agents in interactive decision-making tasks. We introduce Entropy-Regularized Token-level Policy Optimization (ETPO), an entropy-augmented RL method tailored for optimizing LLMs at the token level. We assess the effectiveness of ETPO within a simulated environment that models data science code generation as a series of multi-step interactive tasks.
  arXiv  Detail & Related papers  (2024-02-09T07:45:26Z)

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.