🤖 Transformers - jyunzhang

BioMedGraphica: An All-in-One Platform for Joint Textual Biomedical Prior Knowledge and Numeric Graph Generation

Florian Brand, Prime Intellect research engineer, adopts Gemma 4 E4B 6-bit quantized as his primary local Mac LLM

🤖LLMs News

digg.com··Hacker News

Show HN: LLM memory without context bleed; 100% precision vs. <10% vector search

🤖LLMs

tenureai.dev··Hacker News, Hacker News

Beyond Item IDs: Scaling Short-Form-Video Recommendation via Semantic-Native Long Sequence Modeling

🔍Vector Databases Academic

arxiv.org·

Three sleep intervals for three APIs: Steam 250ms, GitHub 100ms, HuggingFace none

🔌API Design Reference

docs.github.com··DEV

VelocityFM: Short-Horizon Protein Trajectory Prediction via Flow Matching in Velocity Space

⚙️Concurrency Models Academic

biorxiv.org·

techjarves/Portable-AI-USB: A 100% offline, fully portable, zero-trace AI (Ollama + Llama 3 + AnythingLLM) that runs natively from a USB drive on Windows and Mac.

🤖LLMs Code

github.com·

When Vision Misleads, Let Location Speak: A Worldwide Image Geo-Localization Method via Location Attention Mechanism and Large Multimodal Models

📈Optimization Academic

arxiv.org·

google/gemma-4-12B-it-qat-q4_0-gguf

🦙Ollama

huggingface.co·

DUET -- Dual User Embedding Transformers for Offsite Conversion Prediction

📈Optimization Academic

arxiv.org·

Beyond Patches: Superpixel Token-based Transformers for Attribute-Specific Fashion Retrieval

💻Operating Systems Academic

arxiv.org·

princezuda/-RequiemGPT-: Fully open source and open weights built and trained by fable five with one prompt. An experience in how AI actually works

📉Data Science Code

github.com··Hacker News

PENet+: A Lightweight Residual Transformer Framework for Efficient Image Steganalysis

🧠Deep Learning Academic

arxiv.org·

Operator Fusion for LLM Inference on the Tensix Architecture

🤖AI Academic

arxiv.org·

Causal Semantic Alignment for LLM-based Time Series Forecasting

🤖LLMs Academic

arxiv.org·

Audio-Visual Exchange-Aware Token Pruning for Efficient Audio-Visual Captioning

📈Optimization Academic

arxiv.org·

Transformer Based Model for Spatiotemporal Feature Learning in EEG Emotion Recognition

🤖Machine Learning Academic

arxiv.org·

SafeRun: Enabling Determinism in LLM Planning for Running

📝NLP Academic

arxiv.org·

BioMedGraphica: An All-in-One Platform for Joint Textual Biomedical Prior Knowledge and Numeric Graph Generation

Tejas-TA/predikit: The missing bridge between your ML models and your AI agents.

Sakana AI bets AI that improves itself can break the compute arms race of frontier labs

Florian Brand, Prime Intellect research engineer, adopts Gemma 4 E4B 6-bit quantized as his primary local Mac LLM

Show HN: LLM memory without context bleed; 100% precision vs. <10% vector search

Beyond Item IDs: Scaling Short-Form-Video Recommendation via Semantic-Native Long Sequence Modeling

Three sleep intervals for three APIs: Steam 250ms, GitHub 100ms, HuggingFace none

VelocityFM: Short-Horizon Protein Trajectory Prediction via Flow Matching in Velocity Space

techjarves/Portable-AI-USB: A 100% offline, fully portable, zero-trace AI (Ollama + Llama 3 + AnythingLLM) that runs natively from a USB drive on Windows and Mac.

When Vision Misleads, Let Location Speak: A Worldwide Image Geo-Localization Method via Location Attention Mechanism and Large Multimodal Models

google/gemma-4-12B-it-qat-q4_0-gguf

DUET -- Dual User Embedding Transformers for Offsite Conversion Prediction

Beyond Patches: Superpixel Token-based Transformers for Attribute-Specific Fashion Retrieval

princezuda/-RequiemGPT-: Fully open source and open weights built and trained by fable five with one prompt. An experience in how AI actually works

PENet+: A Lightweight Residual Transformer Framework for Efficient Image Steganalysis

Operator Fusion for LLM Inference on the Tensix Architecture

Causal Semantic Alignment for LLM-based Time Series Forecasting

Audio-Visual Exchange-Aware Token Pruning for Efficient Audio-Visual Captioning

Transformer Based Model for Spatiotemporal Feature Learning in EEG Emotion Recognition

SafeRun: Enabling Determinism in LLM Planning for Running