Fwd2Bot: LVLM Visual Token Compression with Double Forward Bottleneck

• URL: http://arxiv.org/abs/2503.21757v1
• Date: Thu, 27 Mar 2025 17:57:07 GMT
• Title: Fwd2Bot: LVLM Visual Token Compression with Double Forward Bottleneck
• Authors: Adrian Bulat, Yassine Ouali, Georgios Tzimiropoulos,
• Abstract summary: We propose a novel compression approach, called Fwd2Bot, that uses the LVLM itself to compress the visual information in a task-agnostic manner.<n>Fwd2Bot results in highly-informative compressed representations suitable for both generative and discriminative tasks.
• Score: 40.21228703978429
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this work, we aim to compress the vision tokens of a Large Vision Language Model (LVLM) into a representation that is simultaneously suitable for (a) generative and (b) discriminative tasks, (c) is nearly lossless, and (d) is storage-efficient. We propose a novel compression approach, called Fwd2Bot, that uses the LVLM itself to compress the visual information in a task-agnostic manner. At the core of Fwd2bot there exists a "double-forward pass" training strategy, whereby, during the first forward pass, the LLM (of the LVLM) creates a bottleneck by condensing the visual information into a small number of summary tokens. Then, using the same LLM, the second forward pass processes the language instruction(s) alongside the summary tokens, used as a direct replacement for the image ones. The training signal is provided by two losses: an autoregressive one applied after the second pass that provides a direct optimization objective for compression, and a contrastive loss, applied after the first pass, that further boosts the representation strength, especially for discriminative tasks. The training is further enhanced by stage-specific adapters. We accompany the proposed method by an in-depth ablation study. Overall, Fwd2Bot results in highly-informative compressed representations suitable for both generative and discriminative tasks. For generative tasks, we offer a 2x higher compression rate without compromising the generative capabilities, setting a new state-of-the-art result. For discriminative tasks, we set a new state-of-the-art on image retrieval and compositionality.

Related papers

• Global Context Compression with Interleaved Vision-Text Transformation [12.971394377165767]
  In this paper, we investigate global context compression, which saves tokens at both prefilling and inference stages.<n>We propose VIST2, a novel Transformer that interleaves input text chunks alongside their visual encoding.<n>With a 4$times$ compression ratio, the resulting models demonstrate significant superiority over baselines on long writing tasks.
  arXiv  Detail & Related papers  (2026-01-15T13:29:16Z)
• CORE: Compact Object-centric REpresentations as a New Paradigm for Token Merging in LVLMs [29.08277140543501]
  We introduce CORE (Compact Object-centric REpresentations), a new paradigm for visual token compression.<n> CORE leverages an efficient segmentation decoder to generate object masks, which serve as a high-level semantic prior to guide the merging of visual tokens.<n>Experiments show that CORE not only establishes a new state-of-the-art on six authoritative benchmarks for fixed-rate compression, but also achieves dramatic efficiency gains in adaptive-rate settings.
  arXiv  Detail & Related papers  (2025-11-18T03:02:23Z)
• Variation-aware Vision Token Dropping for Faster Large Vision-Language Models [24.952668143243542]
  Large vision-language models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding tasks.<n> Token compression offers a direct solution by reducing the number of tokens to be processed, thereby improving computational efficiency.<n>We propose Variation-aware Vision Token Dropping (textiti.e., textbfV$2$Drop), which progressively removes visual tokens with minimal variation during LVLM inference.
  arXiv  Detail & Related papers  (2025-09-01T15:28:44Z)
• Revisit What You See: Disclose Language Prior in Vision Tokens for Efficient Guided Decoding of LVLMs [8.97780713904412]
  This paper introduces ReVisiT, a simple yet effective decoding method that references vision tokens to guide the text generation process in Large Vision-Language Models (LVLMs)<n>Our approach leverages the semantic information embedded within vision tokens by projecting them into the text token distribution space, and dynamically selecting the most relevant vision token at each decoding step through constrained divergence minimization. Experiments on three LVLM benchmarks with two recent LVLMs demonstrate that ReVisiT consistently enhances visual grounding with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-06-11T08:46:55Z)
• Token-Shuffle: Towards High-Resolution Image Generation with Autoregressive Models [92.18057318458528]
  Token-Shuffle is a novel method that reduces the number of image tokens in Transformer. Our strategy requires no additional pretrained text-encoder and enables MLLMs to support extremely high-resolution image synthesis. In GenAI-benchmark, our 2.7B model achieves 0.77 overall score on hard prompts, outperforming AR models LlamaGen by 0.18 and diffusion models LDM by 0.15.
  arXiv  Detail & Related papers  (2025-04-24T17:59:56Z)
• QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation [101.28446308930367]
  Quantized Language-Image Pretraining (QLIP) combines state-of-the-art reconstruction quality with state-of-the-art zero-shot image understanding.<n>QLIP trains a binary-spherical-quantization-based autoencoder with reconstruction and language-image alignment objectives.<n>We demonstrate that QLIP enables a unified mixed-modality auto-regressive model for understanding and generation.
  arXiv  Detail & Related papers  (2025-02-07T18:59:57Z)
• iLLaVA: An Image is Worth Fewer Than 1/3 Input Tokens in Large Multimodal Models [24.0346607116299]
  We introduce iLLaVA, a simple method that can be seamlessly deployed upon current Large Vision-Language Models (LVLMs)<n>iLLaVA achieves this by finding and gradually merging the redundant tokens with an accurate and fast algorithm.<n>On tasks across different domains including single-image, multi-images and videos, iLLaVA demonstrates strong generalizability with consistently promising efficiency.
  arXiv  Detail & Related papers  (2024-12-09T07:22:19Z)
• Looking Beyond Text: Reducing Language bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance [67.26434607115392]
  Large vision-language models (LVLMs) have achieved impressive results in various vision-language tasks. LVLMs suffer from hallucinations caused by language bias, leading to diminished focus on images and ineffective visual comprehension. We propose LACING to address the language bias of LVLMs with muLtimodal duAl-attention meChanIsm (MDA) aNd soft-image Guidance (IFG)
  arXiv  Detail & Related papers  (2024-11-21T16:33:30Z)
• Efficient Large Multi-modal Models via Visual Context Compression [23.966237939194514]
  We present the study on the analysis of redundancy concerning visual tokens and efficient training within large language models. Our initial experiments show that eliminating up to 70% of visual tokens at the testing stage by simply average pooling only leads to a minimal 3% reduction in visual question answering accuracy. We introduce Visual Context on the GQA benchmark, which reduces the number of visual tokens to enhance training and inference efficiency without sacrificing performance.
  arXiv  Detail & Related papers  (2024-06-28T17:57:14Z)
• Vision Model Pre-training on Interleaved Image-Text Data via Latent Compression Learning [78.19528555505961]
  We propose a novel vision model pre-training method called Latent Compression Learning (LCL) for interleaved image-text data.<n>The training objective can be decomposed into two basic tasks: 1) contrastive learning between visual representation and preceding context, and 2) generating subsequent text based on visual representation.<n>Our experiments demonstrate that our method not only matches the performance of CLIP on paired pre-training datasets, but can also leverage interleaved pre-training data.
  arXiv  Detail & Related papers  (2024-06-11T17:59:35Z)
• COTS: Collaborative Two-Stream Vision-Language Pre-Training Model for Cross-Modal Retrieval [59.15034487974549]
  We propose a novel COllaborative Two-Stream vision-language pretraining model termed COTS for image-text retrieval. Our COTS achieves the highest performance among all two-stream methods and comparable performance with 10,800X faster in inference. Importantly, our COTS is also applicable to text-to-video retrieval, yielding new state-ofthe-art on the widely-used MSR-VTT dataset.
  arXiv  Detail & Related papers  (2022-04-15T12:34:47Z)
• IB-DRR: Incremental Learning with Information-Back Discrete Representation Replay [4.8666876477091865]
  Incremental learning aims to enable machine learning models to continuously acquire new knowledge given new classes. Saving a subset of training samples of previously seen classes in the memory and replaying them during new training phases is proven to be an efficient and effective way to fulfil this aim. However, finding a trade-off between the model performance and the number of samples to save for each class is still an open problem for replay-based incremental learning.
  arXiv  Detail & Related papers  (2021-04-21T15:32:11Z)

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.