pith. sign in

explaining to humans everything it knows,

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CL 1

years

2022 1

verdicts

CONDITIONAL 1

representative citing papers

Self-critiquing models for assisting human evaluators

cs.CL · 2022-06-12 · conditional · novelty 6.0

Fine-tuned LLMs produce critiques that improve human detection of errors in summaries, with larger models showing better self-critique and refinement capabilities.

citing papers explorer

Showing 1 of 1 citing paper.

  • Self-critiquing models for assisting human evaluators cs.CL · 2022-06-12 · conditional · none · ref 12

    Fine-tuned LLMs produce critiques that improve human detection of errors in summaries, with larger models showing better self-critique and refinement capabilities.