AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

PromptEmo: Learning Emotion with Bilateral Textual Prompts in Multi-Domain Open-set Scenarios

Xinyi Zeng, Yuxiang Yang, Pinxian Zeng, Wenxia Yin, Bo Liu, Xi Wu, Yan Wang

AAAI 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical alignment and hypothesis testing · 1 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
add positive prompts and positive-oriented CEI losses to the baseline (ablation)PromptEmo · FERPlusAUC-U (%): 58.57 -> 66.64 (+8.07)Improved
add negative prompts and negative-oriented CEI losses (ablation)PromptEmo · FERPlusAUC-U (%): 66.64 -> 69.08 (+2.44)Improved
add explicit semantic separation loss L_ESS (ablation)PromptEmo · FERPlusAUC-U (%): 69.08 -> 69.53 (+0.45)Improved
add latent domain alignment loss L_LDA (ablation)PromptEmo · FERPlusAUC-U (%): 69.53 -> 70.66 (+1.13)Improved
use LLM-generated descriptions instead of 'a photo of {class}' (ablation)PromptEmo positive prompts · FERPlusAUC-U (%): 68.16 -> 70.24 (w/o delta_k)Improved
add trainable prompt vectors delta_k to LLM descriptions (ablation)PromptEmo positive prompts · FERPlusACC-S (%): 56.63 -> 58.04Improved
add a CLIP text encoder with standard prompts to LA-CMFER (ablation)LA-CMFER · FERPlusAUC-U (%): 60.59 -> 60.67No clear effect
PromptEmo vs best prior method (LA-CMFER) on unseen classesMO-FER, four FER datasets · AffectNet/Oulu-CASIA/FERPlus/RAF-DBAUC-U (%): e.g. RAF-DB 68.74 -> 75.91Improved

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