Vision Inference Former: Sustaining Visual Consistency in Multimodal Large Language Models

• URL: http://arxiv.org/abs/2605.18160v1
• Date: Mon, 18 May 2026 10:04:22 GMT
• Title: Vision Inference Former: Sustaining Visual Consistency in Multimodal Large Language Models
• Abstract summary: Vision Inference Former (VIF) is a lightweight architectural module that establishes a bridge between pure visual representations and the model's output space.<n>We conduct experiments on 14 benchmark tasks covering general reasoning, OCR, table understanding, vision-centric evaluation, and hallucination.
• Score: 58.9387276447485
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In recent years, multimodal large language models (MLLMs) have achieved remarkable progress, primarily attributed to effective paradigms for integrating visual and textual information. The dominant connector-based paradigm projects visual features into textual sequence, enabling unified multimodal alignment and reasoning within a generative architecture. However, our experiments reveal two key limitations: (1) Although visual information serves as the core evidential modality in MLLMs, it is treated on par with textual tokens, diminishing the unique contribution of the visual modality; (2) As generation length increases, particularly within a limited context window, the model's dependence on visual information progressively weakens, resulting in deteriorated vision-language alignment and reduced consistency between generated content and visual semantics. To address these challenges, we propose the Vision Inference Former (VIF), a lightweight architectural module that establishes a direct bridge between pure visual representations and the model's output space. Specifically, VIF continuously injects visual semantics throughout the decoding phase of the inference process, ensuring that the model remains firmly grounded in visual content during generation. We conduct experiments on 14 benchmark tasks covering general reasoning, OCR, table understanding, vision-centric evaluation, and hallucination. Experimental results show that VIF consistently improves model performance across diverse architectures while introducing minimal additional overhead. The code for this work is available at https://github.com/Dong-Xinpeng/VIF.

Related papers

• Predictive Regularization Against Visual Representation Degradation in Multimodal Large Language Models [84.94288033791346]
  We conduct a detailed diagnostic analysis to unveil a pervasive issue: visual representation degradation in MLLMs.<n>We attribute this phenomenon to a visual sacrifice driven by the singular text-generation objective, where the model compromises its visual fidelity to optimize for answer generation.<n>We propose Predictive Regularization to force degraded intermediate features to predict initial visual features, thereby maintaining the inherent visual attributes of the MLLM's internal representations.
  arXiv  Detail & Related papers  (2026-03-21T13:10:37Z)
• Stateful Cross-layer Vision Modulation [19.730096071316876]
  multimodal large language models (MLLMs) widely adopt multi-layer visual feature fusion to enhance visual representation.<n>Existing approaches typically perform static concatenation or weighted aggregation after visual encoding, without intervening in the representation formation process itself.<n>We propose a cross-layer memory-modulated vision framework(SCVM) to address these limitations.
  arXiv  Detail & Related papers  (2026-02-28T13:57:19Z)
• AMVICC: A Novel Benchmark for Cross-Modal Failure Mode Profiling for VLMs and IGMs [2.357397994148727]
  multimodal large language models (MLLMs) and image generation models (IGMs) were investigated.<n>We created a novel benchmark to compare failure modes across image-to-text and text-to-image tasks.<n>Our results show that failure modes are often shared between models and modalities, but certain failures are model-specific and modality-specific.
  arXiv  Detail & Related papers  (2026-01-20T00:06:58Z)
• Unleashing the Intrinsic Visual Representation Capability of Multimodal Large Language Models [58.91911788912665]
  We propose Latent Visual Reconstruction (LaVer), a novel training framework that facilitates MLLMs in learning more discrimi visual representations.<n>Our method offers direct visual activation to MLLMs, which exhibit increased visual attention allocation, indicating enhanced utilization of visual information.
  arXiv  Detail & Related papers  (2025-12-06T04:20:13Z)
• v1: Learning to Point Visual Tokens for Multimodal Grounded Reasoning [27.688428439248607]
  We introduce v1, a lightweight extension that enables active visual referencing through a simple point-and-copy approach.<n>This allows the model to identify relevant image patches and copy their embeddings back into the reasoning stream.<n>Our pointing strategy lets the MLLM directly select image patches using their semantic representations as keys, keeping perceptual evidence embedded in the same space as the model's reasoning.
  arXiv  Detail & Related papers  (2025-05-24T19:30:47Z)
• Text Speaks Louder than Vision: ASCII Art Reveals Textual Biases in Vision-Language Models [93.46875303598577]
  Vision-language models (VLMs) have advanced rapidly in processing multimodal information, but their ability to reconcile conflicting signals remains underexplored.<n>This work investigates how VLMs process ASCII art, a unique medium where textual elements collectively form visual patterns, potentially creating semantic-visual conflicts.
  arXiv  Detail & Related papers  (2025-04-02T10:47:07Z)
• VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model [38.61292051733335]
  We present VARGPT, a novel multimodal large language model (MLLM) that unifies visual understanding and generation within a single autoregressive framework.<n>VarGPT employs a next-token prediction paradigm for visual understanding and a next-scale prediction paradigm for visual autoregressive generation.<n> Notably, VARGPT naturally supports capabilities in autoregressive visual generation and instruction-to-image synthesis, showcasing its versatility in both visual understanding and generation tasks.
  arXiv  Detail & Related papers  (2025-01-21T17:50:43Z)
• Unified Language-Vision Pretraining in LLM with Dynamic Discrete Visual Tokenization [52.935150075484074]
  We introduce a well-designed visual tokenizer to translate the non-linguistic image into a sequence of discrete tokens like a foreign language. The resulting visual tokens encompass high-level semantics worthy of a word and also support dynamic sequence length varying from the image. This unification empowers LaVIT to serve as an impressive generalist interface to understand and generate multi-modal content simultaneously.
  arXiv  Detail & Related papers  (2023-09-09T03:01:38Z)
• DiMBERT: Learning Vision-Language Grounded Representations with Disentangled Multimodal-Attention [101.99313208598569]
  Vision-and-language (V-L) tasks require the system to understand both vision content and natural language. We propose DiMBERT (short for Disentangled Multimodal-Attention BERT), which applies separated attention spaces for vision and language. We show that DiMBERT sets new state-of-the-art performance on three tasks.
  arXiv  Detail & Related papers  (2022-10-28T23:00:40Z)

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.