view article Article Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding 12 days ago • 44
TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation Paper • 2602.00268 • Published Jan 30 • 21
AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research Paper • 2507.08038 • Published Jul 9, 2025 • 1
Fast Autoregressive Video Diffusion and World Models with Temporal Cache Compression and Sparse Attention Paper • 2602.01801 • Published Feb 2 • 28
AblationBench Collection This is a collection of datasets used to evaluate language models in the task of ablation planning in empirical AI research. • 5 items • Updated Feb 10 • 6