[KongchangAI]
Benchmark

LEMMO-LongDesc

A long-description subset of the LEMMO corpus, used to evaluate the ability to generate detailed scene-level descriptions of electromagnetic signals.

Timeline (last 90 days)

Nov 9

The research team built and open-sourced a unified electromagnetic signal corpus of 1.38TB, covering real-world and synthetic signals and containing two subsets, LEMMO-ShortQA and LEMMO-LongDesc.

Unverified50%
Nov 9

On the two tasks of LEMMO-LongDesc, LEMMO improved both classification and numerical fact F1.

Unverified50%

All Facts (2)

Source Articles