This is a translation/summary of a Chinese forum post analyzing the paper Alaya-EVOKE: From Linear-Scaling Supervision to Endless World (Yuanyang Yin, Gongxuan Wang, Yifan Zhan, Chuanhao Li, Kaipeng Zhang, Feng Zhao; arXiv: 2608.13546).
Key points
The three curses of interactive world models
- Memory vs. cost: remembering long histories inflates context/KV-cache size; either sessions get truncated or history gets lost — an unavoidable trade-off in naive designs.
- Speed vs. quality: few-step generation quality is capped by its multi-step teacher model.
- Long-term consistency drift: short windows look coherent, but over long rollouts the generated world slowly drifts and violates physical rules.
- Scene geometry is maintained in an external, camera-indexed world state bank rather than inside the model.
- At generation time, only view-relevant state is retrieved — like borrowing a few books from a library instead of carrying them all.
- The denoiser's context stays bounded as sessions grow: memory cost is O(1) regardless of history length (vs. O(T) conventionally).
- Instead of supervising only short segments from real data, the teacher is supervised on sequences it generated itself over 30-second rollouts (multiple chunks, each conditioning on prior generated output).
- This forces the teacher to confront its own error accumulation, ensuring long-horizon consistency — analogous to a writing teacher grading an entire book for coherence, not just individual chapters.
- Three denoising steps per segment, no classifier-free guidance (removing the double forward pass).
- Inherits the bounded-context + recurrent external-memory architecture.
- Performance: 2.11 seconds per 1.5-second clip at 384×640 on a single H200 GPU — near-interactive latency.
- WBench: state-of-the-art.
- VBench-Long: competitive.
- VBench-2.0: competitive.
- The takeaway: EVOKE balances long-horizon consistency, interactive responsiveness, and generation quality rather than maximizing a single metric.
Solution 1: External World State Bank
Solution 2: Teacher redesigned for long-horizon supervision
The teacher uses sparse attention with three components: 1. Chunk-wise grouping — standard attention within chunks (coherence within 'chapters'). 2. Retrieval of selected distant frames — keyframes (scene changes, key events) are explicitly remembered. 3. Linear-attention global state — a compressed global representation capturing long-range dependencies ('the skeleton of the story').This reduces attention cost from O(T²) to O(T), enabling supervision over long-horizon sequences.