# Plug-in Feedback Self-adaptive Attention in CLIP for Training-free Open-Vocabulary Segmentation

arXiv: https://arxiv.org/abs/2508.20265 · HTML: https://mlsanity.com/paper/2508.20265/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 52 of its 67 references that are in the map.
- First posted on arXiv 27 Aug 2025.

## Builds on (most cited first, 30 of 52)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [Semantic Understanding of Scenes through the ADE20K Dataset](https://mlsanity.com/paper/1608.05442/) (2016)
- [COCO-Stuff: Thing and Stuff Classes in Context](https://mlsanity.com/paper/1612.03716/) (2016)
- [Robust fine-tuning of zero-shot models](https://mlsanity.com/paper/2109.01903/) (2021)
- [Extract Free Dense Labels from CLIP](https://mlsanity.com/paper/2112.01071/) (2021)
- [Open-Vocabulary Semantic Segmentation with Mask-adapted CLIP](https://mlsanity.com/paper/2210.04150/) (2022)
- [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://mlsanity.com/paper/2202.11094/) (2022)
- [Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation](https://mlsanity.com/paper/1905.08094/) (2019)
- [RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation](https://mlsanity.com/paper/1611.06612/) (2016)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [SCLIP: Rethinking Self-Attention for Dense Vision-Language Inference](https://mlsanity.com/paper/2312.01597/) (2023)
- [Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs](https://mlsanity.com/paper/2212.00785/) (2022)

## Built on by (most cited first, 12 of 12)

- [Widget2Code: From Visual Widgets to UI Code via Multimodal LLMs](https://mlsanity.com/paper/2512.19918/) (2025)
- [PEARL: Geometry Aligns Semantics for Training-Free Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2603.21528/) (2026)
- [Looking Beyond the Window: Global-Local Aligned CLIP for Training-free Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2603.23030/) (2026)
- [VIP: Visual-guided Prompt Evolution for Efficient Dense Vision-Language Inference](https://mlsanity.com/paper/2605.12325/) (2026)
- [TALON: Test-time Adaptive Learning for On-the-Fly Category Discovery](https://mlsanity.com/paper/2603.08075/) (2026)
- [Test-Time Prototype Adaptation for Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2608.08290/) (2026)
- [Learning through Creation: A Hash-Free Framework for On-the-Fly Category Discovery](https://mlsanity.com/paper/2603.13858/) (2026)
- [SPRITE: From Static Mockups to Engine-Ready Game UI](https://mlsanity.com/paper/2604.18591/) (2026)
- [DouC: Dual-Branch CLIP for Training-Free Open-Vocabulary Segmentation](https://mlsanity.com/paper/2604.24997/) (2026)
- [DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation](https://mlsanity.com/paper/2607.17467/) (2026)
- [SCI-CLIP: Segment-Centric Inference with Reference Memory for Training-Free Open-Vocabulary Segmentation](https://mlsanity.com/paper/2608.05627/) (2026)
- [CARA: Collision-Aware Resolution Adaptation for Multiresolution Hash Encoding Based Image Fitting](https://mlsanity.com/paper/2609.18554/) (2026)

## Abstract (arXiv)

CLIP exhibits strong visual-textual alignment but struggle with open-vocabulary segmentation due to poor localization. Prior methods enhance spatial coherence by modifying intermediate attention. But, this coherence isn't consistently propagated to the final output due to subsequent operations such as projections. Additionally, intermediate attention lacks direct interaction with text representations, such semantic discrepancy limits the full potential of CLIP. In this work, we propose a training-free, feedback-driven self-adaptive framework that adapts output-based patch-level correspondences back to the intermediate attention. The output predictions, being the culmination of the model's processing, encapsulate the most comprehensive visual and textual semantics about each patch. Our approach enhances semantic consistency between internal representations and final predictions by leveraging the model's outputs as a stronger spatial coherence prior. We design key modules, including attention isolation, confidence-based pruning for sparse adaptation, and adaptation ensemble, to effectively feedback the output coherence cues. Our method functions as a plug-in module, seamlessly integrating into four state-of-the-art approaches with three backbones (ViT-B, ViT-L, ViT-H). We further validate our framework across multiple attention types (Q-K, self-self, and Proxy augmented with MAE, SAM, and DINO). Our approach consistently improves their performance across eight benchmarks.