Papers by Kai Konen
Improving Argument Effectiveness Across Ideologies using Instruction-tuned Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a study finds that different political ideologies hold different worldviews, which leads to contentious debates . argument effectiveness is improved by using instruction-tuned large language models . |
| Approach: | They propose to use instruction-tuned large language models to turn ineffective arguments into effective arguments for people with certain ideologies. |
| Outcome: | The proposed methods improve argument effectiveness for liberals by rewriting arguments using three LLM methods. |
Style Vectors for Steering Generative Large Language Models (2024.findings-eacl)
Copied to clipboard
Kai Konen, Sophie Jentzsch, Diaoulé Diallo, Peer Schütt, Oliver Bensch, Roxanne El Baff, Dominik Opitz, Tobias Hecking
| Challenge: | Large language models (LLMs) can be trained on vast corpora and can generate text in a nuanced and parameterisable way. |
| Approach: | They propose to add style vectors to the activations of hidden layers during text generation to steer output towards specific styles. |
| Outcome: | The proposed approach differs from prompt engineering in that it can be nuanced and parameterisable. |