Spaces:
Configuration error
Configuration error
| <div style="max-width:100%;overflow:hidden;"> | |
| <pre style=" | |
| background:#0d0d0d; | |
| font-family:'Courier New',Courier,monospace; | |
| font-size:clamp(3.5px,0.82vw,10px); | |
| line-height:1.15; | |
| padding:0.9em 1.1em; | |
| border-radius:12px; | |
| white-space:pre; | |
| overflow:hidden; | |
| max-width:100%; | |
| box-sizing:border-box; | |
| color:transparent; | |
| background-clip:text; | |
| -webkit-background-clip:text; | |
| background-image:linear-gradient( | |
| 118deg, | |
| #FF6B6B 0%, | |
| #FF8C42 20%, | |
| #FFCF48 40%, | |
| #FFFFFF 55%, | |
| #FFCF48 70%, | |
| #FF8C42 85%, | |
| #FF6B6B 100% | |
| ); | |
| ">βββββββ βββββββ ββββββββ ββββββ ββββ ββββ | |
| βββββββββββββββββββββββββββββββββββββ βββββ | |
| βββ βββββββββββββββββ βββββββββββββββββββ | |
| βββ βββββββββββββββββ βββββββββββββββββββ | |
| βββββββββββ ββββββββββββββ ββββββ βββ βββ | |
| βββββββ βββ ββββββββββββββ ββββββ βββ | |
| ββββ ββββ ββββββ ββββββββββ ββββββββββ βββββββββββ | |
| βββββ ββββββββββββββββββββββββ βββββββββββ βββββββββββ | |
| ββββββββββββββββββββββ βββββββββββββββββ βββββββββ | |
| ββββββββββββββββββββββ βββββββββββββββββββββββββββ | |
| βββ βββ ββββββ ββββββββββββββ βββββββββ ββββββββββββββ | |
| βββ ββββββ βββ ββββββββββ βββββββββ βββββββββββββ</pre> | |
| </div> | |
| <div align="center"> | |
| <p style="font-size:1.6em;font-weight:900;letter-spacing:0.04em;margin:0.2em 0 0.1em;"> | |
| β¨ DreamMachine Research β¨ | |
| </p> | |
| <p style="font-size:1.05em;font-weight:500;color:#aaa;margin:0 0 1em;"> | |
| Billion-Item Neural Recommendation Β· Industrial AI Β· Open Research | |
| </p> | |
| [](https://huggingface.co/dream-machine-ai/Dream-Machine-08-09) | |
| [](https://huggingface.co/datasets/dream-machine-ai/Dream-Machine-08-09-Dataset) | |
| [](https://huggingface.co/spaces/dream-machine-ai/) | |
| [](https://doi.org/10.5281/zenodo.21906715) | |
| [](https://github.com/When-Summer-Understands-Winter/Dream-Machine/) | |
| [](https://creativecommons.org/licenses/by-nc/4.0/) | |
| <br/> | |
| > *"We build the magic shop β the AI that knows exactly what you want before you do."* | |
| **[π€ Models](#-models)** Β· | |
| **[π¦ Datasets](#-datasets)** Β· | |
| **[π Spaces](#-spaces)** Β· | |
| **[π Research](#-research)** Β· | |
| **[βοΈ License](#οΈ-license)** | |
| </div> | |
| --- | |
| ## β¦ Our Mission | |
| ``` | |
| ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ | |
| β β¦ 10ΒΉβ°-item scale β recommend from billion-item catalogues β | |
| β β¦ Open research β weights, data, code, paper all public β | |
| β β¦ Reproducible β every result in the paper can be verified β | |
| β β¦ Production-grade β from laptop to 250 Γ A100 cluster unchanged β | |
| β β¦ Community-first β CC BY-NC 4.0, free for academic use β | |
| ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ | |
| ``` | |
| > **Note on repository location:** All models, datasets, and Spaces are published under the | |
| > **[dream-machine-ai](https://huggingface.co/dream-machine-ai)** account namespace. | |
| > This organization page (`DreamMachine-AI`) serves as the team portal and research hub. | |
| > Follow **[dream-machine-ai](https://huggingface.co/dream-machine-ai)** to get notified of new releases. | |
| --- | |
| ## π€ Models | |
| ### [Dream-Machine-08-09](https://huggingface.co/dream-machine-ai/Dream-Machine-08-09) | |
| > **Billion-Item Neural Recommendation Sorcerer** β Multi-Hash Transformer Dual-Tower with RAG-Enhanced Recall | |
| | Property | Value | | |
| |:---------|:------| | |
| | Architecture | Dual-Tower Transformer (Item + User) | | |
| | Scale | Up to 10ΒΉβ° items, 10ΒΉβ° users | | |
| | Embedding | Multi-Hash K=3, B=50K buckets | | |
| | Layers | L=2 Transformer, H=8 heads, d=128 | | |
| | Retrieval | FAISS HNSW, β€ 100 ms P99 | | |
| | **AUC** | **0.4849** Β· HR@10 **1.0** Β· NDCG@10 **0.537** | | |
| | Format | `safetensors` Β· `transformers`-compatible | | |
| ```python | |
| from transformers import AutoModel | |
| model = AutoModel.from_pretrained( | |
| "dream-machine-ai/Dream-Machine-08-09", trust_remote_code=True | |
| ).eval() | |
| ``` | |
| **[β View Model Card](https://huggingface.co/dream-machine-ai/Dream-Machine-08-09)** | |
| --- | |
| ## π¦ Datasets | |
| ### [Dream-Machine-08-09-Dataset](https://huggingface.co/datasets/dream-machine-ai/Dream-Machine-08-09-Dataset) | |
| > **E-Commerce Synthetic Dataset** β JD / Taobao schema, fully reproducible, zero real PII | |
| | Split | Records | Format | Description | | |
| |:------|--------:|:------:|:------------| | |
| | `products` | 1,000 | JSONL | SKU catalogue β price, brand, category, attributes | | |
| | `users` | 1,000 | JSONL | User profiles β demographics, interests, spending | | |
| | `behaviors` | 5,000 | JSONL | Click / purchase / rating events | | |
| ```python | |
| from datasets import load_dataset | |
| products = load_dataset( | |
| "dream-machine-ai/Dream-Machine-08-09-Dataset", | |
| name="products", split="train" | |
| ) | |
| ``` | |
| **[β View Dataset Card](https://huggingface.co/datasets/dream-machine-ai/Dream-Machine-08-09-Dataset)** | |
| --- | |
| ## π Spaces | |
| ### [dream-machine-ai / README](https://huggingface.co/spaces/dream-machine-ai/README) | |
| Interactive landing page β no installation required. | |
| Explore the model architecture, scaling law charts, and end-to-end deployment guide. | |
| **[β Open Space](https://huggingface.co/spaces/dream-machine-ai/)** | |
| --- | |
| ## π Research | |
| ### DreamMachine: A Billion-Item Neural Recommendation Sorcerer | |
| > Wen, Fangjun Β· DreamMachine Research Team Β· August 2026 | |
| > DOI: [10.5281/zenodo.21906715](https://doi.org/10.5281/zenodo.21906715) | |
| **Seven original contributions:** | |
| | # | Contribution | Key Formula | | |
| |:-:|:-------------|:------------| | |
| | C1 | Multi-Hash Transformer Dual-Tower | Ξ΅_eff = (1 β e^{βN/2B})^K | | |
| | C2 | Hierarchical Feature Architecture | ID β sparse β dense numeric | | |
| | C3 | RAG-Enhanced Multi-Channel Recall | FAISS IVF-PQ + CF + rules | | |
| | C4 | Cross-Attention Listwise Re-ranker | CTR / CVR / dwell-time | | |
| | C5 | Uncertainty-Weighted InfoNCE | Learnable Ο_kΒ² per task | | |
| | C6 | Closed-Form Scaling Laws | 9 breakpoints 10Β² β 10ΒΉβ° | | |
| | C7 | Production Serving Stack | FastAPI + ONNX + INT8 | | |
| ```bibtex | |
| @techreport{wen2026dreammachine, | |
| title = {Magic Shop at Scale: Dream-Machine --- A Billion-Item | |
| Real-Time Neural Recommendation Sorcerer}, | |
| author = {Wen, Fangjun}, | |
| year = {2026}, | |
| institution = {DreamMachine Research Team}, | |
| doi = {10.5281/zenodo.21906715}, | |
| url = {https://doi.org/10.5281/zenodo.21906715}, | |
| note = {Technical Report, August 2026. arXiv cs.IR / cs.LG} | |
| } | |
| ``` | |
| --- | |
| ## π Quick Start | |
| ```bash | |
| pip install transformers torch faiss-cpu huggingface_hub datasets | |
| ``` | |
| ```python | |
| # 1. Load model | |
| from transformers import AutoModel, AutoFeatureExtractor | |
| model = AutoModel.from_pretrained( | |
| "dream-machine-ai/Dream-Machine-08-09", trust_remote_code=True | |
| ).eval() | |
| # 2. Load dataset | |
| from datasets import load_dataset | |
| ds = load_dataset("dream-machine-ai/Dream-Machine-08-09-Dataset", | |
| name="products", split="train") | |
| # 3. Start inference server | |
| # git clone https://huggingface.co/dream-machine-ai/Dream-Machine-08-09 | |
| # python serve_hf.py --model_dir ./hf_model --port 8000 | |
| ``` | |
| --- | |
| ## π All Resources | |
| | Resource | Link | | |
| |:---------|:-----| | |
| | π€ **Model** | [dream-machine-ai/Dream-Machine-08-09](https://huggingface.co/dream-machine-ai/Dream-Machine-08-09) | | |
| | π¦ **Dataset** | [dream-machine-ai/Dream-Machine-08-09-Dataset](https://huggingface.co/datasets/dream-machine-ai/Dream-Machine-08-09-Dataset) | | |
| | π **Space** | [spaces/dream-machine-ai](https://huggingface.co/spaces/dream-machine-ai/) | | |
| | π’ **Org** | [huggingface.co/DreamMachine-AI](https://huggingface.co/DreamMachine-AI) | | |
| | π **Paper** | [Zenodo 10.5281/zenodo.21906715](https://doi.org/10.5281/zenodo.21906715) | | |
| | π» **GitHub** | [When-Summer-Understands-Winter/Dream-Machine](https://github.com/When-Summer-Understands-Winter/Dream-Machine/) | | |
| --- | |
| ## βοΈ License | |
| All models, datasets, and code published by DreamMachine Research are licensed under the | |
| **Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0)**. | |
| | β Permitted | β Prohibited | | |
| |:---|:---| | |
| | Academic research & publication | Commercial products or services | | |
| | Personal learning & experimentation | Revenue-generating deployments | | |
| | Non-commercial derivative works | Sublicensing for profit | | |
| | Citing in papers with attribution | Any business use without written permission | | |
| > **Commercial use of any kind β including integrating model weights, code, or outputs into a product or service β requires explicit prior written consent from the author (Fangjun Wen).** | |
| > Contact: fangjunwen168@outlook.com | |
| [](https://creativecommons.org/licenses/by-nc/4.0/) | |
| --- | |
| <div align="center"> | |
| <p style="font-size:1.05em;font-weight:700;margin:0.5em 0 0.3em;"> | |
| π€ <a href="https://huggingface.co/dream-machine-ai/Dream-Machine-08-09">Model</a> Β· | |
| π¦ <a href="https://huggingface.co/datasets/dream-machine-ai/Dream-Machine-08-09-Dataset">Dataset</a> Β· | |
| π <a href="https://huggingface.co/spaces/dream-machine-ai/">Space</a> Β· | |
| π <a href="https://doi.org/10.5281/zenodo.21906715">Paper</a> Β· | |
| π» <a href="https://github.com/When-Summer-Understands-Winter/Dream-Machine/">GitHub</a> | |
| </p> | |
| **DreamMachine Research Team** Β· Fangjun Wen Β· August 2026 | |
| `// built with focus Β· DreamMachine Research` | |
| </div> | |