🧊 3D Vision

💬 ACL2025 · 1 paper notes

📌 Same area in other venues: 📷 CVPR2026 (751) · 🔬 ICLR2026 (197) · 🧪 ICML2026 (30) · 🤖 AAAI2026 (79) · 🧠 NeurIPS2025 (116) · 📹 ICCV2025 (267)

Slamming: Training a Speech Language Model on One GPU in a Day

This paper proposes the Slam training recipe, which systematically optimizes model initialization, architectural choices, synthetic data, and preference alignment to train a speech language model on a single A5000 GPU within 24 hours, achieving performance comparable to large-scale SLMs.