Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DiffuSent: Towards a Unified Diffusion Framework for Aspect-Based Sentiment Analysis

About

Aspect-Based Sentiment Analysis (ABSA) encompasses seven distinct subtasks, each focusing on different extracted elements. Despite the proven success of generative models in unified aspect sentiment analysis, existing approaches often rely on auto-regressive token-by-token generation without grasping the whole information of the aspect and opinion terms, resulting in boundary insensitivity, particularly in context of multi-word aspect and opinion terms. To address these issues, we present DiffuSent, a non-auto-regressive diffusion framework that systematically formulates all ABSA subtasks as boundary denoising diffusion processes, progressively refining boundaries over noisy states. Furthermore, we introduce a contrastive denoising training strategy which effectively address duplicate predictions with subtle variations introduced by diffusion process. Extensive experiments across 28 settings (7 subtasks x 4 datasets) demonstrate that DiffuSent achieves delivers consistent improvements over the strongest generative and span-based systems. DiffuSent exhibits notable gains on multi-word triplets, achieving an average improvement of +2.48 F1, and maintains robust extraction accuracy in sentences containing multiple sentiment triplets. Moreover, the non-auto-regressive decoding enables substantial efficiency benefits, reaching up to 181 times faster inference than auto-regressive generative baselines

Shu Long, Yanglei Gan, Xuchuan Zhou• 2026

Related benchmarks

TaskDatasetResultRank
aspect sentiment triplet extractionRes15 D20b (test)
F1 Score66.39
15
aspect sentiment triplet extractionRes16 D20b (test)
F1-score74.22
15
aspect sentiment triplet extractionLap14 D20b (test)
F1 Score63.03
15
aspect sentiment triplet extractionRes14 D20b (test)
F1-score74.97
14
Aspect Sentiment Triplet Extraction (ASTE)D20a Lap14
F1 Score63.31
12
Aspect Sentiment Triplet Extraction (ASTE)D20a Res14
F1 Score74.91
12
Aspect Sentiment Triplet Extraction (ASTE)D20a Res15
F1-score67.58
12
Aspect Sentiment Triplet Extraction (ASTE)D20a Res16
F1-score75.09
12
Aspect-Opinion Pair Extraction (AOPE)D20a Lap14
F1-score71.67
8
Aspect-Opinion Pair Extraction (AOPE)D20a Res14
F1 Score79.86
8
Showing 10 of 16 rows

Other info

Follow for update