For GPT-3.5-turbo and GPT-4, we incorporate ETF and ICL following our proposed method for fair comparison

Computational complexity, F1 score comparison of LLM-based methods Method Computational complexity F1 score (%) Sampling Training Inference SciERC JNLPBA BC5CDR GPT-3 · 2024

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

1 Pith paper citing it

browse 1 citing papers

representative citing papers

Enhancing Scientific Named Entity Recognition via Large Language Models: A Type-driven Multi-task Learning Approach

cs.CL · 2026-08-09

citing papers explorer

Showing 1 of 1 citing paper.

Enhancing Scientific Named Entity Recognition via Large Language Models: A Type-driven Multi-task Learning Approach cs.CL · 2026-08-09 · unreviewed · ref 10