📄 文章列表
42 篇文章
TDS
Learning Triton One Kernel at a Time: Softmax
TDS
Neural Networks Are Blurry, Symbolic Systems Are Fragmented. Sparse Autoencoders Help Us Combine Them.
TDS
Training Objective That Makes LLM Inference 3X Faster
TDS
NeurIPS 2025 Best Paper Review: Qwen’s Systematic Exploration of Attention Gating
TDS
Optimizing Data Transfer in Batched AI/ML Inference Workloads
TDS
Cutting LLM Memory by 84%: A Deep Dive into Fused Kernels
TDS
How Convolutional Neural Networks Learn Musical Similarity
TDS
Distributed Reinforcement Learning for Scalable High-Performance Policy Optimization
TDS
Mechanistic Interpretability: Peeking Inside an LLM
TDS
Optimizing Token Generation in PyTorch Decoder Models
TDS
Hybrid Neuro-Symbolic Fraud Detection: Guiding Neural Networks with Domain Rules
TDS
Hallucinations in LLMs Are Not a Bug in the Data
TDS
How a Neural Network Learned Its Own Fraud Rules: A Neuro-Symbolic AI Experiment
TDS
Detecting Translation Hallucinations with Attention Misalignment
TDS
A Guide to Voice Cloning on Voxtral with a Missing Encoder
TDS
Introduction to Deep Evidential Regression for Uncertainty Quantification
TDS
Bytes Speak All Languages: Cross-Script Name Retrieval via Contrastive Learning
TDS
From Possible to Probable AI Models
TDS
The AI Model Confidence Trap
TDS