Organizations: Independent Researcher · Centre for Applied Creative Technologies (CFACT+), Bournemouth University, UK · School of Computing, Newcastle University, Newcastle upon Tyne, UK
Dimensional Aspect-Based Sentiment Analysis (DimABSA) extends traditional ABSA from categorical polarity labels to continuous valence-arousal (VA) regression. This paper describes a system developed for Track A, Subtask 1 (Dimensional Aspect Sentiment Regression), aiming to predict real-valued VA scores in the [1, 9] range for each given aspect in a text. A fine-tuning approach based on XLM-RoBERTa-base is adopted, with dual regression heads with sigmoid-scaled outputs for valence and arousal prediction. Separate models are trained for each language-domain pair (English and Chinese across restaurant, laptop, and finance domains), and training and development sets are merged for final test predictions. In development experiments, the fine-tuning approach is compared against several large language models under a few-shot prompting setting, demonstrating that task-specific fine-tuning outperforms these LLM-based methods across all evaluation datasets.
Figures & tables
Figure 1: Model architecture.
Language
Domain
Train
Dev
Test
eng
Restaurant
2,284
200
1,000
eng
Laptop
4,076
200
1,000
zho
Restaurant
6,050
300
1,000
zho
Laptop
3,490
300
1,000
zho
Finance
1,000
200
842
Table 1: Dataset statistics (number of sentences).
Hyperparameter
Value
Pretrained model
XLM-RoBERTa-base
Max sequence length
256
Batch size
16
Learning rate
2e-5
Optimizer
AdamW
Warmup ratio
10%
Table 2: Hyperparameters for fine-tuning.
Method
eng_lap
eng_res
zho_lap
zho_res
zho_fin
Kimi-K2 Thinking
2.1893
2.1461
1.6440
1.8959
1.9652
Qwen-3 14B (QLoRA)
2.8089
2.6427
1.7706
2.0073
1.4707
XLM-RoBERTa
1.4562
1.4861
0.7510
0.9553
0.5391
Table 3: Test set RMSE VA compared with baselines (lower is better).
Figure 2: Validation RMSE VA across training epochs.
Method
eng_lap
eng_res
zho_lap
zho_res
zho_fin
GPT-5.2
1.9663
1.9203
1.5825
1.7279
1.7311
LLaMA-3-70B
1.6243
1.6748
1.9249
1.9589
2.2312
LLaMA-4-Mav.
1.6987
1.7955
1.6480
1.7718
1.7858
XLM-RoBERTa
1.1084
1.3402
0.8034
0.6727
0.5118
Table 4: RMSE VA comparison on the 20% held-out evaluation set (lower is better).
Figure 3: Visual comparison of RMSE VA across methods on the 20% held-out evaluation set.
Figure 4: Training data distribution in VA space. English datasets cluster in the high-valence, high-arousal region, while Chinese datasets are more centered.
Figure 5: Test-set RMSE VA heatmap across the VA space. Each cell shows the RMSE and sample count (n) for gold samples in that valence–arousal bin. Red indicates higher error; green indicates lower error.
Language
Domain
RMSE V
RMSE A
eng
Laptop
1.25
0.75
eng
Restaurant
1.11
0.98
zho
Laptop
0.57
0.48
zho
Restaurant
0.61
0.74
zho
Finance
0.44
0.31
Table 5: Test-set RMSE for valence (RMSE V ) and arousal (RMSE A ).