MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer Paper • 2509.16197 • Published Sep 19, 2025 • 57
Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces Paper • 2601.11868 • Published 12 days ago • 30
PACEvolve: Enabling Long-Horizon Progress-Aware Consistent Evolution Paper • 2601.10657 • Published 13 days ago • 20
Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs Paper • 2510.18245 • Published Oct 21, 2025 • 7