Fine-tuning NLLB-200 for a New Low-Resource Language in 2026
FormosanBank
• • 1NEW Articles from Team or Enterprise organizations will get promoted to the main section.
GOLDENER <br> Smart batching from embeddings and clustering
Yann-CV
• Give your agents ZeroGPU to ship viral AI apps autonomously
victor
• • 2Qwen3-0.6B Uni-Queue-Klassifikator (DE) — Rein aus Metadaten trainiert
Tobi-Bueck
• • 1Writing High-Performance Kernels in TileLang, from GEMM to MLA
AtlasCloud-AI
• ToricGT: noncommutative torus memory for graph-token reasoning
AmelieSchreiber
• Should we use genetics instead of system prompts for AI Agents & Personas?
nyxia
• • 2Borealis — open data, code, weights recipe for training Audio LLM
AlexWortega
• • 14Offline ASR Benchmark on Call-Center Audio
norwooodsystems
• Lucky Defaults: A Failure Mode in Probe-Based Generalisation Metrics
ymode
• Relaunching PapersWithCode with new features
nielsr
• • 8LLM Guardrail Models are Less Robust Against Text Mutation Attacks
kalyan-ks
• • 1An experiment with attention.
wop
• • 5📄 Research Papers Dataset
tegridydev
• • 2Eight Days in China: What I Learned from the AI Labs, Robotics Startups and Academia
matthew-d-white
• • 3Text Degeneration: A Production Failure Mode That Most Benchmarks Do Not Track
Dharma-AI
• • 1Why Open Models Are the Only Sustainable Way to Teach AI
penelopegittos
• • 7Scaling Zero-Resource Vocabulary: A Data Pipeline for Sango
MEYNG
• When does autoresearch need a human?
ProlificAI
• LeRobot Humanoid: An Open, Low-Cost, 3D-Printed Humanoid for Robot Learning
VirgileBatto
• • 41