Can Feedback Enhance Semantic Grounding in Large Vision-Language Models?

• URL: http://arxiv.org/abs/2404.06510v1
• Date: Tue, 9 Apr 2024 17:59:04 GMT
• Title: Can Feedback Enhance Semantic Grounding in Large Vision-Language Models?
• Authors: Yuan-Hong Liao, Rafid Mahmood, Sanja Fidler, David Acuna,
• Abstract summary: We investigate whether Vision-Language Models (VLMs) can improve their semantic grounding by "receiving" feedback. We find that if prompted appropriately, VLMs can utilize feedback both in a single step and iteratively. We show grounding accuracy consistently improves using automated feedback across all models in all settings investigated.
• Score: 61.899791071654654
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Enhancing semantic grounding abilities in Vision-Language Models (VLMs) often involves collecting domain-specific training data, refining the network architectures, or modifying the training recipes. In this work, we venture into an orthogonal direction and explore whether VLMs can improve their semantic grounding by "receiving" feedback, without requiring in-domain data, fine-tuning, or modifications to the network architectures. We systematically analyze this hypothesis using a feedback mechanism composed of a binary signal. We find that if prompted appropriately, VLMs can utilize feedback both in a single step and iteratively, showcasing the potential of feedback as an alternative technique to improve grounding in internet-scale VLMs. Furthermore, VLMs, like LLMs, struggle to self-correct errors out-of-the-box. However, we find that this issue can be mitigated via a binary verification mechanism. Finally, we explore the potential and limitations of amalgamating these findings and applying them iteratively to automatically enhance VLMs' grounding performance, showing grounding accuracy consistently improves using automated feedback across all models in all settings investigated. Overall, our iterative framework improves semantic grounding in VLMs by more than 15 accuracy points under noise-free feedback and up to 5 accuracy points under a simple automated binary verification mechanism. The project website is hosted at https://andrewliao11.github.io/vlms_feedback

Related papers

• Locality Alignment Improves Vision-Language Models [55.275235524659905]
  Vision language models (VLMs) have seen growing adoption in recent years, but many still struggle with basic spatial reasoning errors. We propose a new efficient post-training stage for ViTs called locality alignment. We show that locality-aligned backbones improve performance across a range of benchmarks.
  arXiv  Detail & Related papers  (2024-10-14T21:01:01Z)
• Learning to Ground VLMs without Forgetting [54.033346088090674]
  We introduce LynX, a framework that equips pretrained Visual Language Models with visual grounding ability without forgetting their existing image and language understanding skills. To train the model effectively, we generate a high-quality synthetic dataset we call SCouT, which mimics human reasoning in visual grounding. We evaluate LynX on several object detection and visual grounding datasets, demonstrating strong performance in object detection, zero-shot localization and grounded reasoning.
  arXiv  Detail & Related papers  (2024-10-14T13:35:47Z)
• Repairs in a Block World: A New Benchmark for Handling User Corrections with Multi-Modal Language Models [48.42142115255159]
  We release BlockWorld-Repairs: a dataset of multi-modal TPR sequences in an instruction-following manipulation task. We evaluate several state-of-the-art Vision and Language Models (VLM) across multiple settings, focusing on their capability to process and accurately respond to TPRs. Our results suggest that these models are not yet ready to be deployed in multi-modal collaborative settings.
  arXiv  Detail & Related papers  (2024-09-21T21:06:25Z)
• MarvelOVD: Marrying Object Recognition and Vision-Language Models for Robust Open-Vocabulary Object Detection [107.15164718585666]
  We investigate the root cause of VLMs' biased prediction under the open vocabulary detection context. Our observations lead to a simple yet effective paradigm, coded MarvelOVD, that generates significantly better training targets. Our method outperforms the other state-of-the-arts by significant margins.
  arXiv  Detail & Related papers  (2024-07-31T09:23:57Z)
• Improving the Validity of Automatically Generated Feedback via Reinforcement Learning [50.067342343957876]
  We propose a framework for feedback generation that optimize both correctness and alignment using reinforcement learning (RL) Specifically, we use GPT-4's annotations to create preferences over feedback pairs in an augmented dataset for training via direct preference optimization (DPO)
  arXiv  Detail & Related papers  (2024-03-02T20:25:50Z)
• Distribution-Aware Prompt Tuning for Vision-Language Models [20.02599087680773]
  A key to prompt tuning is the feature space alignment between two modalities via learnable vectors with model parameters fixed. Inspired by this observation, we proposed distribution-aware prompt tuning (DAPT) for vision-language models. Our experiments on 11 benchmark datasets demonstrate that our method significantly improves generalizability.
  arXiv  Detail & Related papers  (2023-09-06T23:49:11Z)
• Can I Trust Your Answer? Visually Grounded Video Question Answering [88.11169242115416]
  We study visually grounded VideoQA in response to the emerging trends of utilizing pretraining techniques for video-language understanding. We construct NExT-GQA -- an extension of NExT-QA with 10.5$K$ temporal grounding labels tied to the original QA pairs.
  arXiv  Detail & Related papers  (2023-09-04T03:06:04Z)
• ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases [16.308432111311195]
  Vision Transformers (ViTs) rely on more flexible self-attention layers, and have recently outperformed CNNs for image classification. We introduce gated positional self-attention (GPSA), a form of positional self-attention which can be equipped with a "soft" convolutional inductive bias. The resulting convolutional-like ViT architecture, ConViT, outperforms the DeiT on ImageNet, while offering a much improved sample efficiency.
  arXiv  Detail & Related papers  (2021-03-19T09:11:20Z)

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.