AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Eliciting Positive Emotion through Affect-Sensitive Dialogue Response Generation: A Neural Network Approach

Nurul Lubis, Sakriani Sakti, Koichiro Yoshino, Satoshi Nakamura

AAAI 2018 · Large language models › Language, safety and interpretability · approach: Faithfulness of LLM explanations · 79 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune only emotion encoder and utterance decoder (selective) instead of all parameters (ablation)HRED · pos_sem testperplexity: 121.44 -> 100.94Improved
selective fine-tuning with sem data (ablation)HRED · pos_sem testperplexity: 185.66 -> 151.77Improved
add emotion encoder (Emo-HRED vs HRED), selective, pos sem (ablation)HRED · pos_sem testperplexity: 100.94 -> 51.34Improved
add emotion encoder, selective, sem (ablation)HRED · pos_sem testperplexity: 151.77 -> 69.66Improved
fine-tune on positive-emotion eliciting responses (pos sem) instead of sem (ablation)Emo-HRED · pos_sem testperplexity: 69.66 -> 51.34Improved
add emotion encoder, pos sem data (ablation)HRED · Human evalnaturalness: 3.26 -> 3.59Improved
add emotion encoder, pos sem data (ablation)HRED · Human evalemotional impact: 3.22 -> 3.58Improved
fine-tune on pos sem instead of sem (ablation)HRED · Human evalemotional impact: 2.56 -> 3.22Improved
fine-tune on pos sem instead of sem (ablation)HRED · Human evalnaturalness: 2.71 -> 3.26Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy