PulseAugur
EN
LIVE 21:00:53

Goku dataset and benchmark advance instruction-based video editing

By PulseAugur Editorial · [3 sources] ·

Researchers have introduced Goku, a new dataset and benchmark designed for instruction-based video editing. Goku comprises 2 million video editing pairs, expanding beyond simple appearance edits to include complex multi-task and structural manipulations. The accompanying Goku-Edit model, which utilizes a multimodal large language model for instruction comprehension, demonstrates an improvement of up to 8% over existing open-source models on the newly proposed Goku-Bench benchmark. AI

IMPACT Advances capabilities in instruction-based video editing, potentially enabling more complex and creative video manipulation tools.

RANK_REASON The cluster describes a new academic paper introducing a dataset, benchmark, and model for video editing.

Read on Hugging Face Daily Papers →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

Goku dataset and benchmark advance instruction-based video editing

COVERAGE [3]

  1. Hugging Face Daily Papers TIER_1 English(EN) ·

    Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing

    A large-scale video editing dataset and model are introduced that support multi-task and structural manipulations through advanced data synthesis and network architectures.

  2. arXiv cs.CV TIER_1 English(EN) · Sen Liang, Cong Wang, Zhentao Yu, Fengbin Guan, Zhengguang Zhou, Teng Hu, Youliang Zhang, Yuan Zhou, Xin Li, Qinglin Lu, Zhibo Chen ·

    Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing

    arXiv:2606.30599v1 Announce Type: new Abstract: Existing instruction-based video editing datasets commonly focus on single-task appearance editing, failing to meet the complex creative demands of real-world scenarios. To bridge this gap, we present Goku, a large-scale dataset fea…

  3. arXiv cs.CV TIER_1 English(EN) · Zhibo Chen ·

    Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing

    Existing instruction-based video editing datasets commonly focus on single-task appearance editing, failing to meet the complex creative demands of real-world scenarios. To bridge this gap, we present Goku, a large-scale dataset featuring 2 million high-quality, instruction-align…

RELATED ENTITIES

RELATED TOPICS