Hugging Face's logo

Spaces:
hyper3labs
/
README
Running

App Files Community
Matin Mahmood commited on
Commit
2aa5ac7
·
1 Parent(s): b6a844a

Refresh organization profile and branding

Browse files
Files changed (2) hide show
  1. README.md +24 -35
  2. logo.svg +0 -0
README.md CHANGED
@@ -1,57 +1,46 @@
1
  ---
2
  title: README
3
  emoji: 🌐
4
- colorFrom: red
5
- colorTo: purple
6
  sdk: static
7
  pinned: false
8
  ---
9
 
10
  <div align="center">
11
- <img src="logo.svg" width="160" alt="hyper³labs logo" style="margin-bottom: 20px; border-radius: 12px;" />
12
  <h1>hyper³labs</h1>
13
- <p style="font-size: 1.2rem; color: #888; margin-top: -10px;">
14
- <i>Your data is hierarchical. Your models should be too.</i>
15
- </p>
16
  <p>
17
- <a href="https://hyper3labs.github.io"><img src="https://img.shields.io/badge/Website-hyper3labs.github.io-blue?style=flat-square&logo=google-chrome&logoColor=white&color=007A87" alt="Website" /></a>
18
- <a href="https://github.com/hyper3labs"><img src="https://img.shields.io/badge/GitHub-hyper3labs-black?style=flat-square&logo=github&logoColor=white&color=181717" alt="GitHub" /></a>
19
  </p>
20
  </div>
21
 
22
- ---
23
-
24
- Welcome to **hyper³labs** (pronounced *Hypercube Labs*) — an applied research and engineering lab building competitive embedding models and tools to inspect, curate, and improve embedding and retrieval systems.
25
-
26
- We believe that **Euclidean space should be a choice, not a default**. We develop representations and visualization methods that respect the intrinsic geometry of data, particularly hierarchical, taxonomic, and non-Euclidean structures.
27
 
28
- ## 🔮 Interactive Workbenches (Spaces)
29
 
30
- We develop **HyperView**, an open-source workbench for understanding embedding spaces and dataset/model failures. Explore our live interactive demos:
31
 
32
- * **[🔮 HyperView Main Demo](https://huggingface.co/spaces/hyper3labs/HyperView)** — Our core open-source workbench for visualizing high-dimensional embedding spaces, mapping taxonomy, and diagnosing model/dataset failures.
33
- * **[🐆 Jaguar Spot Re-ID](https://huggingface.co/spaces/hyper3labs/HyperView-Jaguar-ReID)** — Spot-pattern diagnostic visualization for wildlife conservation. Part of our collaboration on individual identification and diagnostic verification.
34
- * **[📦 ABO Catalog Explorer](https://huggingface.co/spaces/hyper3labs/HyperView-ABO-Catalog)** — Visualizing taxonomies and fine-grained product hierarchies using the Amazon Berkeley Objects dataset.
35
- * **[🔍 VisA Industrial Search](https://huggingface.co/spaces/hyper3labs/HyperView-VisA-Manufacturing)** — Anomaly inspection and structural part finding in industrial manufacturing datasets.
36
- * **[👗 DeepFashion Search](https://huggingface.co/spaces/hyper3labs/HyperView-DeepFashion-Text-Search)** — Exploring zero-shot retrieval, attribute mapping, and structural hierarchies in e-commerce fashion datasets.
37
- * **[🎯 Precision Region Search](https://huggingface.co/spaces/hyper3labs/precision-region-search-refcocog-hyper3clip)** — Interactive referring expression grounding demo using RefCOCOg and the Hyper3-CLIP model.
38
 
39
- ## 🧠 Open Models
40
 
41
- * **[🤗 hyper3labs/hyper3-clip-v0.5](https://huggingface.co/hyper3labs/hyper3-clip-v0.5)** — Our open-weight 512-dimensional Lorentzian (hyperbolic) vision-language checkpoint trained with compositional entailment constraints for hierarchy-sensitive retrieval.
 
 
 
 
 
42
 
43
- ## 📊 Datasets
44
 
45
- * **[🤗 hyper3labs/jaguar-re-id](https://huggingface.co/datasets/hyper3labs/jaguar-re-id)** — Spot patterns dataset used for individual jaguar identification in the wild.
46
- * **[🤗 hyper3labs/amazon-berkeley-objects](https://huggingface.co/datasets/hyper3labs/amazon-berkeley-objects)** — Mirror of the Amazon Berkeley Objects dataset used for fine-grained hierarchical product search evaluation.
47
- * **[🤗 hyper3labs/jaguar-hyperview-demo](https://huggingface.co/datasets/hyper3labs/jaguar-hyperview-demo)** — Stratified subset for spot-pattern verification and debugging.
48
 
49
- ## 📄 Publications
50
-
51
- * **Are We Recognizing the Jaguar or Its Background? A Diagnostic Framework for Jaguar Re-Identification**
52
- *Antonio Rueda-Toicen, Robert Vava, Matin Mahmood.* arXiv:2604.09690 (April 2026).
53
- [arXiv:2604.09690](https://arxiv.org/abs/2604.09690) | [Hugging Face Dataset](https://huggingface.co/datasets/hyper3labs/jaguar-re-id)
54
-
55
- ---
56
 
57
- *Building the geometry of intelligence.*
 
 
1
  ---
2
  title: README
3
  emoji: 🌐
4
+ colorFrom: blue
5
+ colorTo: gray
6
  sdk: static
7
  pinned: false
8
  ---
9
 
10
  <div align="center">
11
+ <img src="logo.svg" width="150" alt="hyper³labs logo" />
12
  <h1>hyper³labs</h1>
13
+ <p><strong>Embedding models and retrieval infrastructure.</strong></p>
 
 
14
  <p>
15
+ <a href="https://hyper3labs.com">Website</a> ·
16
+ <a href="https://github.com/Hyper3Labs">GitHub</a>
17
  </p>
18
  </div>
19
 
20
+ hyper³labs (pronounced *Hypercube Labs*) builds embedding models and retrieval infrastructure across text and vision. We also build open-source tools for evaluating and understanding embedding systems.
 
 
 
 
21
 
22
+ We are especially interested in how hierarchy and non-Euclidean geometry affect retrieval, and in the failures that aggregate benchmark scores hide.
23
 
24
+ ## Models
25
 
26
+ - **[hyper3-clip-v0.5](https://huggingface.co/hyper3labs/hyper3-clip-v0.5)** is an open-weight vision-language embedding model for hierarchy-sensitive retrieval. It produces 512-dimensional embeddings and works with the Sentence Transformers interface.
 
 
 
 
 
27
 
28
+ ## Tools and demos
29
 
30
+ - **[HyperView](https://huggingface.co/spaces/hyper3labs/HyperView)** is our open-source workbench for exploring embedding spaces and tracing retrieval failures back to real samples.
31
+ - **[ABO Catalog Explorer](https://huggingface.co/spaces/hyper3labs/HyperView-ABO-Catalog)** tests fine-grained retrieval on product catalogs.
32
+ - **[DeepFashion Search](https://huggingface.co/spaces/hyper3labs/HyperView-DeepFashion-Text-Search)** explores text-to-image search for garments and product attributes.
33
+ - **[Precision Region Search](https://huggingface.co/spaces/hyper3labs/precision-region-search-refcocog-hyper3clip)** demonstrates referring-expression retrieval on RefCOCOg.
34
+ - **[Jaguar Re-ID](https://huggingface.co/spaces/hyper3labs/HyperView-Jaguar-ReID)** examines whether re-identification models recognize an animal or its background.
35
+ - **[VisA Industrial Search](https://huggingface.co/spaces/hyper3labs/HyperView-VisA-Manufacturing)** explores anomaly and part retrieval in industrial imagery.
36
 
37
+ ## Datasets
38
 
39
+ - **[jaguar-re-id](https://huggingface.co/datasets/hyper3labs/jaguar-re-id)**: individual jaguar identification from spot patterns.
40
+ - **[amazon-berkeley-objects](https://huggingface.co/datasets/hyper3labs/amazon-berkeley-objects)**: a mirror used for hierarchical product retrieval evaluation.
41
+ - **[jaguar-hyperview-demo](https://huggingface.co/datasets/hyper3labs/jaguar-hyperview-demo)**: a smaller subset prepared for interactive analysis.
42
 
43
+ ## Research
 
 
 
 
 
 
44
 
45
+ **[Are We Recognizing the Jaguar or Its Background? A Diagnostic Framework for Jaguar Re-Identification](https://arxiv.org/abs/2604.09690)**
46
+ Antonio Rueda-Toicen, Robert Vava, Matin Mahmood. April 2026.
logo.svg CHANGED