Mastodon discussion Discussions Sep 17 3 views

A new benchmark evaluates open-source instruction-tuned LLMs for key-value pair extraction under clean-text and noisy OC...

by Sipirtu

A new benchmark evaluates open-source instruction-tuned LLMs for key-value pair extraction under clean-text and noisy OCR conditions. Results show performance degrades substantially under OCR noise.Source: arXiv cs.CLhttps://arxiv.org/abs/2609.17538#MachineLearning

Read Original
Benchmark Open Source

Metadata

Account
sipirtu

Related