Unverified50% confidenceFactExact time

LatComm trains only the cross-attention reading module and does not fine-tune the encoder and receiver.

1

Sources

50%

Confidence

Long-term

Relevance

10/20/2026

First Seen

Sources

LatComm: Reusable Context Compression Speeds Up LLM-to-LLM Communication by 3x

Related Entities

LatCommconcept Cross-Attentionconcept

Related Claims

UnverifiedThe decoder adds two mechanisms on top of the encoder: masked self-attention and cross-attention; masked self-attention ensures that during generation only previously produced tokens can be seen, while cross-attention allows the decoder to query the encoder's output.63% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/1073263

API

curl https://kongchang.com/api/v1/knowledge/claims/1073263

MCP

get_claim(id=1073263)