Hugging Face – Posts

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

All HF Hub posts

Nymbo

posted an update 1 day ago

Post

3771

We should really have a release date range slider on the /models page. Tired of "trending/most downloaded" being the best way to sort and still seeing models from 2023 on the first page just because they're embedded in enterprise pipelines and get downloaded repeatedly. "Recently Created/Recently Updated" don't solve the discovery problem considering the amount of noise to sift through.

Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.

3 replies

OzTianlu

posted an update 2 days ago

Post

5215

Arcade-3B — SmolReasoner
NoesisLab/Arcade-3B
Arcade-3B is a 3B instruction-following and reasoning model built on SmolLM3-3B. It is the public release from the ARCADE project at NoesisLab, which investigates the State–Constraint Orthogonality Hypothesis: standard Transformer hidden states conflate factual content and reasoning structure in the same subspace, and explicitly decoupling them improves generalization.

5 replies

prithivMLmods

posted an update 3 days ago

Post

4737

QIE-2509-Object-Remover-Bbox-v3 is a more stable version of the Qwen Image Edit visual grounding–based object removal model. The app was previously featured in HF Spaces of the Week and is now updated with the latest Bbox-v3 LoRA adapter.

🤗 Demo: prithivMLmods/QIE-Object-Remover-Bbox
🤗 LoRA: prithivMLmods/QIE-2509-Object-Remover-Bbox-v3
🤗 Collection: https://huggingface.co/collections/prithivMLmods/qwen-image-edit-layout-bbox

To learn more, visit the app page or the respective model pages.

2 replies

unmodeled-tyler

posted an update 1 day ago

Post

3023

LINK: https://github.com/unmodeled-tyler/vessel-browser

Hey Hugging Face!

It's been quiet from me over here for the last few weeks, but I've been busy building! I just submitted my project to the Hermes Agent Hackathon, and wanted to share it with all of you.

This is Vessel Browser - an AI-native web browser that runs locally on Linux, and is operated by your personal AI agent via MCP server. Vessel is built from the ground up around the agent as first-class and visible UI for human-in-the-loop with 3 different levels of permissions.

Your agent finds, reads, and organizes the web for you, based on what you actually care about - not what a platform's algorithm thinks you care about.

Once your agent finds what it's looking for, it can organize bookmarked pages into custom folders with summaries for later browsing, take screenshots with highlighted text, and integrate with Obsidian for long-term browsing related-memory.

Check it out!

1 reply

BibbyResearch

posted an update 3 days ago

Post

2444

We are working on the largest Dataset and Pre-trained model for Text to Speech and Speech to text for the low-resourced language called Marwari in India.

danielhanchen

posted an update 3 days ago

Post

3532

We collaborated with NVIDIA to teach you about Reinforcement Learning and RL environments. 💚 Learn:

• Why RL environments matter + how to build them
• When RL is better than SFT
• GRPO and RL best practices
• How verifiable rewards and RLVR work

Blog: https://unsloth.ai/blog/rl-environments

4 replies

robtacconelli

posted an update about 14 hours ago

Post

149

🧬 Midicoth: diffusion-based lossless compression — no neural net, no GPU, no training data

What if reverse diffusion could compress text — without a neural network?
Midicoth brings score-based denoising into classical compression. It treats prior smoothing as forward noise and reverses it with Tweedie's formula on a binary tree — 3 denoising steps, James-Stein shrinkage, applied after all model blending. ~2,000 lines of C, single CPU core.

Beats every dictionary compressor we tested:
enwik8 (100 MB) → 1.753 bpb (−11.9% vs xz, −15% vs Brotli, −24.5% vs bzip2)
alice29.txt → 2.119 bpb (−16.9% vs xz)
Outperforms xz, zstd, Brotli, bzip2, gzip on all inputs

PAQ/CMIX still win with hundreds of models + LSTMs. LLM compressors win with pre-trained knowledge. Midicoth closes the gap with pure statistics — no mixer, no gradient descent, just counting.
The Tweedie denoising layer adds 2.3–2.7% on every file tested — the most consistent component in the ablation. Adding SSE or logistic mixers made things worse. In the online setting, count-based beats gradient-based.
No external dependencies. Fully deterministic. Bit-exact encode/decode. ~60 KB/s throughput.
💻 Code: https://github.com/robtacconelli/midicoth
📄 Paper: Micro-Diffusion Compression -- Binary Tree Tweedie Denoising for Online Probability Estimation (2603.08771)
⭐ Space: robtacconelli/midicoth

If you ever wondered whether diffusion ideas belong in data compression — here's proof they do. ⭐ appreciated!

kanaria007

posted an update about 21 hours ago

Post

117

✅ Article highlight: *Federated SI* (art-60-044, v0.1)

TL;DR:
Most real systems do not live inside a single SI-Core. Cities, hospital networks, grid operators, transit systems, vendors, and neighboring institutions all run under different governance, trust, and legal boundaries.

This note sketches *Federated SI*: how multiple SI-Cores coordinate without pretending to share one brain. The focus is on portable artifacts, explicit trust boundaries, negotiated goals, limited memory exchange, and graceful failure when cooperation partially breaks.

Read:
kanaria007/agi-structural-intelligence-protocols

Why it matters:
• makes cross-operator coordination explicit instead of hiding it inside ad hoc APIs
• supports cooperation under separate trust anchors, legal regimes, and policy surfaces
• treats failure modes seriously: partitions, vetoes, degraded cooperation, partial visibility
• keeps governance portable via normalized verdicts, pinned bindings, and export-safe artifacts

What’s inside:
• why “one SI-Core sees everything” is the wrong default
• federation objects such as federated SIRs, goal surfaces, memory views, and consent records
• negotiation across cities, hospitals, utilities, and other institutional stacks
• operational labels vs exported governance verdicts (ACCEPT / DEGRADE / REJECT)
• deterministic, auditable exchange rules for cross-run / cross-vendor comparison
• failover, mutual aid, and graceful degradation when trust or connectivity breaks

Key idea:
Intelligence at institution scale is not a single runtime. It is a *federation of governed runtimes* that must negotiate, coordinate, and fail safely without collapsing auditability.

W8Yi

posted an update 1 day ago

Post

I built a **TCGA WSI feature dataset using UNI2-h**.

The official release currently has incomplete coverage (see discussion):
MahmoodLab/UNI2-h-features#2

To make the features easier to use for research, I generated a new dataset:

W8Yi/tcga-wsi-uni2h-features

Key differences from the official release:

• **All detected tissue tiles are encoded** (not a sampled subset)
• **Features can be downloaded per slide** instead of large ZIP archives
• **QC overlay images** are provided for visual inspection
• **UNI2-h 1536-D tile embeddings** stored in H5 format
• Organized by TCGA project for easier use in MIL / retrieval pipelines

Example layout:

TCGA-HNSC/
  features/*.h5
  vis/*__overlay.png

Hope this helps others working on computational pathology and TCGA WSI research.

karstenskyt

posted an update about 5 hours ago

Post

🚀 𝗪𝗲 𝗷𝘂𝘀𝘁 𝗼𝗽𝗲𝗻𝗲𝗱 𝘁𝗵𝗲 𝗱𝗼𝗼𝗿𝘀 𝘁𝗼 𝗹𝘂𝘅𝘂𝗿𝘆-𝗹𝗮𝗸𝗲𝗵𝗼𝘂𝘀𝗲! Our full 12-page open-source soccer analytics platform is now live and publicly accessible on Hugging Face Spaces.

We successfully migrated our entire backend Streamlit dashboard to a Docker Space. Instead of being locked behind workspace accounts, anyone can now directly explore our models, 38M+ tracking frames, and advanced metrics (xT, PAUSA, DEFCON) in the browser.

To prepare for the public release, we ran a massive optimization sprint (what used to be days of normal engineering, now condensed into a few hours of "Claude Time"):

⚡ 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲: Eliminated O(n²) bottlenecks, dropping ingestion pipeline runs from 36 minutes to <30 seconds.

🧠 𝗨𝗫: Applied a deep Cognitive Interface Audit across all surfaces to ensure the data is intuitive and human-readable.

The Hugging Face ecosystem now natively handles our entire public-facing presentation layer—Models, Datasets, Gradio Demos, and the full Streamlit App.

𝗧𝗿𝘆 𝘁𝗵𝗲 𝗹𝗶𝘃𝗲 𝗮𝗽𝗽: luxury-lakehouse/soccer-analytics-app

𝗘𝘅𝗽𝗹𝗼𝗿𝗲 𝘁𝗵𝗲 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲:

luxury-lakehouse

Recently active users