Skip to content
Tagged articles

VBench

3 articles · Page 1 of 1
AI Frontier Lectures
AI Frontier Lectures
Nov 3, 2026 · Artificial Intelligence

Wave Forcing Triples Long-Video Motion Quality Over Rolling Forcing

Shanghai Jiao Tong University's DENG Lab introduces Wave Forcing, a pipelined denoising schedule that raises 100-second video dynamics from 36.63 to 91.45 on VBench, achieves 81.64 overall score, and delivers 7x throughput on 8 H200 GPUs while preserving subject consistency at 97.10.

H200 GPUVBenchWan2.1
0 likes · 6 min read
Wave Forcing Triples Long-Video Motion Quality Over Rolling Forcing
AIWalker
AIWalker
Mar 31, 2025 · Artificial Intelligence

VBench-2.0: A Next‑Generation Benchmark for Intrinsic Faithfulness in AI Video Generation

VBench-2.0 expands the original VBench suite by introducing six fine‑grained dimensions—Human Fidelity, Controllability, Creativity, Physics, Commonsense, and more—to evaluate not only the visual quality of generated videos but also their intrinsic faithfulness to physical laws, common sense, and narrative coherence, providing open‑source tools, prompts, and human‑aligned metrics for the research community.

AI-evaluationBenchmarkIntrinsic Faithfulness
0 likes · 12 min read
VBench-2.0: A Next‑Generation Benchmark for Intrinsic Faithfulness in AI Video Generation
NewBeeNLP
NewBeeNLP
Mar 14, 2025 · Artificial Intelligence

How Open‑Sora 2.0 Achieves SOTA Video Generation with Only $200K Training Cost

Open‑Sora 2.0 is an open‑source 11B‑parameter video generation model that matches commercial SOTA performance while being trained on 224 GPUs for just $200,000, thanks to a 3D auto‑encoder, MMDiT architecture, aggressive data filtering, low‑resolution pre‑training, and highly optimized parallel training techniques.

AI ModelMMDiTOpen-Sora
0 likes · 9 min read
How Open‑Sora 2.0 Achieves SOTA Video Generation with Only $200K Training Cost