Elevating Visual Perception in Multimodal LLMs with Visual Embedding Distillation

• URL: http://arxiv.org/abs/2412.09585v3
• Date: Fri, 17 Oct 2025 00:33:41 GMT
• Title: Elevating Visual Perception in Multimodal LLMs with Visual Embedding Distillation
• Authors: Jitesh Jain, Zhengyuan Yang, Humphrey Shi, Jianfeng Gao, Jianwei Yang,
• Abstract summary: In recent times, the standard practice for developing MLLMs is to feed features from vision encoder(s) into the LLM and train with natural language supervision.<n>This approach often causes models to lean towards language comprehension and undermine the rich visual perception signals present in the data.<n>We propose VisPer-LM, the first approach that infuses visual perception knowledge from expert vision encoders into the LLM's hidden representations.
• Score: 109.5893580175657
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In recent times, the standard practice for developing MLLMs is to feed features from vision encoder(s) into the LLM and train with natural language supervision. This approach often causes models to lean towards language comprehension and undermine the rich visual perception signals present in the data, which are critical for tasks involving spatial reasoning in the domain of embodied AI and robotics. Is it possible to optimize both at the same time? In this work, we propose VisPer-LM, the first approach that infuses visual perception knowledge from expert vision encoders into the LLM's (of an MLLM) hidden representations. We start by investigating MLLMs trained solely with natural language supervision and identify a positive correlation between the quality of visual representations within these models and their downstream performance. Given this insight, we formulate the objective during the pretraining stage in MLLMs as a coupled optimization of predictive visual embedding and next (text) token prediction. Moreover, through extensive probing, we observe improved visual representation quality due to embedding optimization, underscoring the effectiveness of our probing setup. We demonstrate that our VisPer-LM outperforms the single and multi-encoder baselines, proving our approach's superiority over explicitly feeding the corresponding features to the LLM. In particular, VisPer-LM boosts performance by an average margin of up to 2.5% on various benchmarks, with a notable improvement of 8.7% on the Depth task in CV-Bench.

Related papers

• RL makes MLLMs see better than SFT [96.508432109136]
  We conduct a critical yet under-explored analysis of the vision encoder of Multimodal Language Model (MLLM)<n>Our results demonstrate that MLLM's post-training strategy (i.e., SFT or RL) not only leads to distinct outcomes on MLLM downstream tasks, but also fundamentally reshapes MLLM's underlying visual representations.<n>We then reframe our findings into a simple recipe for building strong vision encoders for MLLMs, Preference-Instructed Vision OpTimization (PIVOT)
  arXiv  Detail & Related papers  (2025-10-18T03:37:17Z)
• Vision-Centric Activation and Coordination for Multimodal Large Language Models [42.26911585599856]
  Multimodal large language models (MLLMs) integrate image features from visual encoders with LLMs, demonstrating advanced comprehension capabilities.<n>However, mainstream MLLMs are solely supervised by the next-token prediction of textual tokens, neglecting critical vision-centric information.<n>We introduce VaCo, which optimize MLLM representations through Vision-Centric activation and Coordination.
  arXiv  Detail & Related papers  (2025-10-16T06:38:39Z)
• VIS-Shepherd: Constructing Critic for LLM-based Data Visualization Generation [17.6462454905092]
  We introduce VIS-Shepherd, a specialized Multimodal Large Language Model (MLLM)-based critic.<n>At the core of our approach is a framework to construct a high-quality visualization critique dataset.<n>Our experiments show that even small (7B parameters) open-source MLLM models achieve substantial performance gains.
  arXiv  Detail & Related papers  (2025-06-16T10:15:38Z)
• Expanding the Boundaries of Vision Prior Knowledge in Multi-modal Large Language Models [51.84752285423123]
  We introduce a novel metric, $Rank_e$, to quantify the effect of prior knowledge of the vision encoder on MLLM performance.<n>We propose VisPRE (Vision Prior Remediation), a two-stage training framework that explicitly incorporates prior knowledge at the vision encoder level.<n> Experimental results demonstrate that augmenting vision encoder's prior knowledge substantially boosts the visual understanding capabilities of MLLMs.
  arXiv  Detail & Related papers  (2025-03-23T11:33:09Z)
• AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Understanding [63.09928907734156]
  AlignVLM is a vision-text alignment method that maps visual features to a weighted average of text embeddings. Our experiments show that AlignVLM achieves state-of-the-art performance compared to prior alignment methods.
  arXiv  Detail & Related papers  (2025-02-03T13:34:51Z)
• AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning [19.68349294206012]
  We propose a training-free adaptive inference method for multi-modal LLMs. With a minimalist design, our method can be applied to both video and image LLMs. Under a similar computational cost, our method outperforms the state-of-the-art methods in long video understanding.
  arXiv  Detail & Related papers  (2024-12-04T11:47:57Z)
• Enhancing Perception Capabilities of Multimodal LLMs with Training-Free Fusion [40.56646959926701]
  Multimodal LLMs (MLLMs) equip language models with visual capabilities by aligning vision encoders with language models.<n>Existing methods to enhance the visual perception of MLLMs often involve designing more powerful vision encoders.<n>We introduce VisionFuse, a novel integration framework that efficiently utilizes multiple vision encoders from off-the-shelf MLLMs.
  arXiv  Detail & Related papers  (2024-12-02T09:02:28Z)
• Mono-InternVL: Pushing the Boundaries of Monolithic Multimodal Large Language Models with Endogenous Visual Pre-training [48.455597568212944]
  We present Mono-InternVL, a novel monolithic MLLM that seamlessly integrates a set of visual experts via a multimodal mixture-of-experts structure. In particular, EViP is designed as a progressive learning process for visual experts, which aims to fully exploit the visual knowledge from noisy data to high-quality data.
  arXiv  Detail & Related papers  (2024-10-10T17:59:22Z)
• GLOV: Guided Large Language Models as Implicit Optimizers for Vision Language Models [44.82179903133343]
  Large Language Models (LLMs) act as implicits for Vision-Langugage Models (VLMs)<n>Our GLOV meta-prompts an LLM with the downstream task description, querying it for suitable VLM prompts.<n>We evaluate our GLOV on 16 diverse datasets using two families of VLMs.
  arXiv  Detail & Related papers  (2024-10-08T15:55:40Z)
• Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs [61.143381152739046]
  We introduce Cambrian-1, a family of multimodal LLMs (MLLMs) designed with a vision-centric approach.<n>Our study uses LLMs and visual instruction tuning as an interface to evaluate various visual representations.<n>We provide model weights, code, supporting tools, datasets, and detailed instruction-tuning and evaluation recipes.
  arXiv  Detail & Related papers  (2024-06-24T17:59:42Z)
• Dense Connector for MLLMs [89.50595155217108]
  We introduce the Dense Connector - a plug-and-play vision-language connector that significantly enhances existing MLLMs. Building on this, we also propose the Efficient Dense Connector, which achieves performance comparable to LLaVA-v1.5 with only 25% of the visual tokens. Our model, trained solely on images, showcases remarkable zero-shot capabilities in video understanding as well.
  arXiv  Detail & Related papers  (2024-05-22T16:25:03Z)
• Mitigating Object Hallucination in Large Vision-Language Models via Image-Grounded Guidance [51.30560006045442]
  Image-gRounded guIdaNcE (MARINE) is a framework that is both training-free and API-free.<n>MARINE effectively and efficiently reduces object hallucinations during inference by introducing image-grounded guidance to LVLMs.<n>Our framework's flexibility further allows for the integration of multiple vision models, enabling more reliable and robust object-level guidance.
  arXiv  Detail & Related papers  (2024-02-13T18:59:05Z)
• LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge [58.82222646803248]
  Multimodal Large Language Models (MLLMs) have endowed LLMs with the ability to perceive and understand multi-modal signals. Most of the existing MLLMs mainly adopt vision encoders pretrained on coarsely aligned image-text pairs, leading to insufficient extraction and reasoning of visual knowledge. We propose a dual-Level vIsual knedgeOwl eNhanced Multimodal Large Language Model (LION), which empowers the MLLM by injecting visual knowledge in two levels.
  arXiv  Detail & Related papers  (2023-11-20T15:56:44Z)
• InfMLLM: A Unified Framework for Visual-Language Tasks [44.29407348046122]
  multimodal large language models (MLLMs) have attracted growing interest. This work delves into enabling LLMs to tackle more vision-language-related tasks. InfMLLM achieves either state-of-the-art (SOTA) performance or performance comparable to recent MLLMs.
  arXiv  Detail & Related papers  (2023-11-12T09:58:16Z)
• Sight Beyond Text: Multi-Modal Training Enhances LLMs in Truthfulness and Ethics [32.123919380959485]
  Multi-modal large language models (MLLMs) are trained based on large language models (LLM) While they excel in multi-modal tasks, the pure NLP abilities of MLLMs are often underestimated and left untested. We show that visual instruction tuning, a prevailing strategy for transitioning LLMs into MLLMs, unexpectedly and interestingly helps models attain both improved truthfulness and ethical alignment.
  arXiv  Detail & Related papers  (2023-09-13T17:57:21Z)
• On Learning to Summarize with Large Language Models as References [101.79795027550959]
  Large language models (LLMs) are favored by human annotators over the original reference summaries in commonly used summarization datasets. We study an LLM-as-reference learning setting for smaller text summarization models to investigate whether their performance can be substantially improved.
  arXiv  Detail & Related papers  (2023-05-23T16:56:04Z)

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.