The authors studied multi-agent workflows arranged as directed graphs, including parallel fan-in tasks on FanOutQA and sequential communication on LongBench v2.
Receiver-specific cache sharing cuts multi-agent costs while improving accuracy
CacheBack uses a receiving agent’s request to select relevant sender states, avoiding transfer of the full attention cache.
Big Tech
Maximillian Rossi · Prajwal Raghunath · Haoqing Xuan · Yusen Zhang · Eugene Wu
Columbia University · Amazon Web Services
Research Digest··2 min read
Rossi et al.
Why this paper
From Amazon Web Services and Columbia University
In one line
Receiver-conditioned KV-cache filtering cuts multi-agent communication state while improving accuracy and latency over text messaging across several model architectures.
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 (2 benchmarks)
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§