The authors introduce Chain Dimension Scaling (CDS), which repeatedly applies one feature-interaction block using the same parameters at every iteration.
Reusing Ranking Blocks Cuts Communication While Preserving Short-Video Quality
Chain Dimension Scaling repeatedly applies a shared feature-interaction block, increasing effective computation depth without adding trainable parameters.
Big Tech
Zhimeng Jiang · Yuening Li · Ruixiao Sun · Yuanzhen Lin · Kun Jin · Chuan He · +1 more
Research Digest··2 min read
Jiang and colleagues tested parameter-shared iterative computation in a production, multi-task short-video ranking system.
Why this paper
From Google
In one line
CDS reduces parameters by 26.6% and improves training throughput by 6.4% in a production ranking system.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§