Binding Visual Features Point by Point

• URL: http://arxiv.org/abs/2605.25427v1
• Date: Mon, 25 May 2026 04:58:23 GMT
• Title: Binding Visual Features Point by Point
• Abstract summary: We show that learning to point-via-text induces an internal visual search routine.<n>We also find that pointing behavior can be generalized to new tasks via fine-tuning.
• Score: 6.577814524465175
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite success on standard benchmarks, vision language models display persistent failures on tasks involving processing of multi-object scenes, including many tasks that are relatively easy for humans. Recent work has found that these failures may stem from a basic inability to accurately bind object features in-context, a challenge that is referred to as the "binding problem" in cognitive science and neuroscience. The human visual system is thought to solve this binding problem via serial processing, attending to individual objects one at a time so as to avoid interference from other objects. Recent work has proposed "pointing" -- the use of explicit spatial coordinates to refer to objects -- as an analogous solution for vision language models, and found that it improves performance on challenging multi-object tasks. However, it is unclear $\textit{why}$ (i.e., on a mechanistic or representational level) this approach improves performance, and how directly this relates to serial processing in human vision. Here, we investigate this question. We find that learning to point-via-text induces an internal visual search routine, and we characterize the mechanisms that support this procedure. We also find that pointing behavior can be generalized to new tasks via fine-tuning, and that doing so eliminates binding errors and enables compositional generalization. These results provide a proof-of-principle that serial processing can solve the binding problem for vision language models just as it does for biological vision.

Related papers

• Can Unified Generation and Understanding Models Maintain Semantic Equivalence Across Different Output Modalities? [61.533560295383786]
  Unified Multimodal Large Language Models (U-MLLMs) integrate understanding and generation within a single architecture.<n>We observe that U-MLLMs fail to maintain semantic equivalence when required to render the same results in the image modality.<n>We introduce VGUBench, a framework to decouple reasoning logic from generation fidelity.
  arXiv  Detail & Related papers  (2026-02-27T06:23:56Z)
• Latent Implicit Visual Reasoning [59.39913238320798]
  We propose a task-agnostic mechanism that trains LMMs to discover and use visual reasoning tokens without explicit supervision.<n>Our approach outperforms direct fine-tuning and achieves state-of-the-art results on a diverse range of vision-centric tasks.
  arXiv  Detail & Related papers  (2025-12-24T14:59:49Z)
• Weakly-Supervised Learning of Dense Functional Correspondences [23.794395724229762]
  We propose a weakly-supervised learning paradigm to tackle the prediction task.<n>The main insight behind our approach is that we can leverage vision-language models to obtain functional parts.<n>We then integrate this with dense contrastive learning from pixel correspondences to distill both functional and spatial knowledge into a new model.
  arXiv  Detail & Related papers  (2025-09-04T05:39:16Z)
• Do VLMs Have Bad Eyes? Diagnosing Compositional Failures via Mechanistic Interpretability [2.1432646818975014]
  Vision-Language Models (VLMs) have shown remarkable performance in integrating visual and textual information for tasks such as image captioning and visual question answering.<n>These models struggle with compositional generalization and object binding.<n>Our work explores the root causes of these failures using mechanistic interpretability techniques.
  arXiv  Detail & Related papers  (2025-08-20T01:15:28Z)
• ObjectRelator: Enabling Cross-View Object Relation Understanding Across Ego-Centric and Exo-Centric Perspectives [109.11714588441511]
  The Ego-Exo object correspondence task aims to understand object relations across ego-exo perspectives through segmentation.<n> PSALM, a recently proposed segmentation method, stands out as a notable exception with its demonstrated zero-shot ability on this task.<n>We propose ObjectRelator, a novel approach featuring two key modules: Multimodal Condition Fusion and SSL-based Cross-View Object Alignment.
  arXiv  Detail & Related papers  (2024-11-28T12:01:03Z)
• Understanding the Limits of Vision Language Models Through the Lens of the Binding Problem [37.27516441519387]
  We show that state-of-the-art vision language models exhibit surprising failures on basic multi-object reasoning tasks that humans perform with near perfect accuracy.<n>We find that many of the puzzling failures of state-of-the-art VLMs can be explained as arising due to the binding problem, and that these failure modes are strikingly similar to the limitations exhibited by rapid, feedforward processing in the human brain.
  arXiv  Detail & Related papers  (2024-10-31T22:24:47Z)
• OLIVE: Object Level In-Context Visual Embeddings [8.168219870640318]
  We propose a novel method to prompt large language models with in-context visual object vectors. This eliminates the necessity of fusing a lengthy array of image patch features and significantly speeds up training. Our experiments reveal that our method achieves competitive referring object classification and captioning performance.
  arXiv  Detail & Related papers  (2024-06-02T21:36:31Z)
• CogCoM: A Visual Language Model with Chain-of-Manipulations Reasoning [61.21923643289266]
  Chain of Manipulations is a mechanism that enables Vision-Language Models to solve problems step-by-step with evidence.<n>After training, models can solve various visual problems by eliciting intrinsic manipulations (e.g., grounding, zoom in) actively without involving external tools.<n>Our trained model, textbfCogCoM, achieves state-of-the-art performance across 9 benchmarks from 4 categories.
  arXiv  Detail & Related papers  (2024-02-06T18:43:48Z)
• Few-Shot In-Context Imitation Learning via Implicit Graph Alignment [15.215659641228655]
  We formulate imitation learning as a conditional alignment problem between graph representations of objects. We show that this conditioning allows for in-context learning, where a robot can perform a task on a set of new objects immediately after the demonstrations.
  arXiv  Detail & Related papers  (2023-10-18T18:26:01Z)
• Look, Remember and Reason: Grounded reasoning in videos with language models [5.3445140425713245]
  Multi-temporal language models (LM) have recently shown promising performance in high-level reasoning tasks on videos. We propose training an LM end-to-end on low-level surrogate tasks, including object detection, re-identification, tracking, to endow the model with the required low-level visual capabilities. We demonstrate the effectiveness of our framework on diverse visual reasoning tasks from the ACRE, CATER, Something-Else and STAR datasets.
  arXiv  Detail & Related papers  (2023-06-30T16:31:14Z)
• Images Speak in Images: A Generalist Painter for In-Context Visual Learning [98.78475432114595]
  In-context learning allows the model to rapidly adapt to various tasks with only a handful of prompts and examples. It is unclear how to define the general-purpose task prompts that the vision model can understand and transfer to out-of-domain tasks. We present Painter, a generalist model which redefines the output of core vision tasks as images, and specify task prompts as also images.
  arXiv  Detail & Related papers  (2022-12-05T18:59:50Z)
• INVIGORATE: Interactive Visual Grounding and Grasping in Clutter [56.00554240240515]
  INVIGORATE is a robot system that interacts with human through natural language and grasps a specified object in clutter. We train separate neural networks for object detection, for visual grounding, for question generation, and for OBR detection and grasping. We build a partially observable Markov decision process (POMDP) that integrates the learned neural network modules.
  arXiv  Detail & Related papers  (2021-08-25T07:35:21Z)
• Exploit Clues from Views: Self-Supervised and Regularized Learning for Multiview Object Recognition [66.87417785210772]
  This work investigates the problem of multiview self-supervised learning (MV-SSL) A novel surrogate task for self-supervised learning is proposed by pursuing "object invariant" representation. Experiments shows that the recognition and retrieval results using view invariant prototype embedding (VISPE) outperform other self-supervised learning methods.
  arXiv  Detail & Related papers  (2020-03-28T07:06:06Z)

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.