QuadAI at SemEval-2026 Task 3: Ensemble Learning of Hybrid RoBERTa and LLMs for Dimensional Aspect-Based Sentiment Analysis

A. J. W. de Vink; Filippos Karolos Ventirozos; Natalia Amat-Lefort; Lifeng Han

cs.CLMar 8, 2026

QuadAI at SemEval-2026 Task 3: Ensemble Learning of Hybrid RoBERTa and LLMs for Dimensional Aspect-Based Sentiment Analysis

Authors: A. J. W. de Vink, Filippos Karolos Ventirozos, Natalia Amat-Lefort, Lifeng Han

Organizations: LIACS, Leiden University, NL · Manchester Metropolitan University, UK · Leiden University Medical Center, NL

Abstract

We present our system for SemEval-2026 Task 3 on dimensional aspect-based sentiment regression. Our approach combines a hybrid RoBERTa encoder, which jointly predicts sentiment using regression and discretized classification heads, with large language models (LLMs) via prediction-level ensemble learning. The hybrid encoder improves prediction stability by combining continuous and discretized sentiment representations. We further explore in-context learning with LLMs and ridge-regression stacking to combine encoder and LLM predictions. Experimental results on the development set show that ensemble learning significantly improves performance over individual models, achieving substantial reductions in RMSE and improvements in correlation scores. Our findings demonstrate the complementary strengths of encoder-based and LLM-based approaches for dimensional sentiment analysis. Our development code and resources will be shared at https://github.com/aaronlifenghan/ABSentiment

Figures & tables

Figure 1: Hybrid RoBERTa

Figure 2: Triple-LLMs workflow

Panel of Figure 3 — Figure 3: Prediction-level ensemble architecture combining base models and optional VADER features.

Model Variant	MSE $\downarrow$	RMSE $\downarrow$	RMSE v $\downarrow$	RMSE a $\downarrow$	$\rho_{v}$ $\uparrow$	$\rho_{a}$ $\uparrow$	$\rho_{mean}$ $\uparrow$
Regression only	0.6140	0.7836	0.7201	0.8423	0.9102	0.5505	0.7304
Bin (expected value)	0.6238	0.7898	0.8340	0.7430	0.8974	0.5126	0.7050
Hybrid (average)	0.5419	0.7361	0.7214	0.7506	0.9074	0.5388	0.7231

Table 1: Performance of the Hybrid RoBERTa model on SemEval Task 1 ( Laptop , Dev set). The model combines a regression head and a discretized classification head (31 bins), with the final prediction obtained by averaging both outputs (

w=0.5

). Error catergories: lower is better; in 3 catergories hyrid is much better, except for RMSE that is comparable to Bin. Pearson correation

\rho

higher is better rho( Regress>Hybrid>Bin)

$num\_bins$	$\alpha$	$w$	RMSE avg $\downarrow$	$\rho_{mean}$ $\uparrow$	$\rho_{a}$ $\uparrow$
31	0.5	0.5	0.7361	0.7231	0.5388
11	1.0	0.5	0.7368	0.7182	0.5277
31	0.5	0.4	0.7376	0.7252	0.5421
7	0.2	0.5	0.7405	0.7287	0.5473
31	0.5	0.3	0.7431	0.7270	0.5448
11	1.0	0.4	0.7448	0.7185	0.5278

Table 2: Top-10 Hybrid configurations ranked by average RMSE on the Laptop Dev set.

num\_bins

denotes the number of discretization bins,

\alpha

the classification loss weight, and

w

the regression–classification averaging weight.

Domain	Variant	MSE $\downarrow$	RMSE $\downarrow$	RMSE v $\downarrow$	RMSE a $\downarrow$	$\rho_{v}$ $\uparrow$	$\rho_{a}$ $\uparrow$	$\rho_{mean}$ $\uparrow$
Restaurant	Regression	0.8176	0.9042	0.8212	0.9802	0.9205	0.6458	0.7832
	Bin-expected	0.5369	0.7327	0.8154	0.6395	0.9130	0.6408	0.7769
	Average ( $w=0.5$ )	0.4919	0.7013	0.6692	0.7320	0.9217	0.6679	0.7948

Table 3: Hybrid RoBERTa results on the Restaurant development set for SemEval Task 1 (Valence/Arousal). The model combines a regression head and a hard-bin classification head; the average prediction is a weighted combination with

\texttt{pred\_weight}=0.5

. Best configuration and detailed dev metrics are shown.

Setting	Method	MSE $\downarrow$	RMSE $\downarrow$	RMSE v $\downarrow$	RMSE a $\downarrow$	$\rho_{v}$ $\uparrow$	$\rho_{a}$ $\uparrow$	$\rho_{mean}$ $\uparrow$
Without VADER	Avg	0.4095	0.6399	0.5835	0.6918	0.9391	0.6105	0.7748
	Weighted (w=[0.3,0.7])	0.4025	0.6344	0.5629	0.6987	0.9421	0.6124	0.7773
	Stacking (Ridge, OOF)	0.4025	0.6344	0.5713	0.6918	0.9397	0.5893	0.7645
With VADER	Avg	0.4095	0.6399	0.5835	0.6918	0.9391	0.6105	0.7748
	Weighted (w=[0.3,0.7])	0.4025	0.6344	0.5629	0.6987	0.9421	0.6124	0.7773
	Stacking (Ridge, OOF)	0.4079	0.6387	0.5718	0.6992	0.9396	0.5793	0.7594

Table 4: Ensemble results on the Laptop Dev set for SemEval Task 1 (Valence/Arousal). We compare simple averaging, weighted averaging, and ridge-regression stacking (out-of-fold, OOF) with and without additional VADER-based features. Best (lowest) RMSE among valid (non-leaking) settings is highlighted.

Best config	num_bins	$\alpha_{\text{cls}}$	pred_weight ( $w$ )
	21	1.0	0.3

Table 5: Track B (English) Environmental Protection — best hybrid configuration and DEV results.

Model	MSE	RMSE	RMSE v	RMSE a	$\rho$
LLM (ICL)	0.484	0.695	0.633	0.752	0.757

Table 6: LLM dev results (Laptop).

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 4: Example of TaskA1 data

Figure 5: Example of TaskB data

Figure 7: English Restaurant Data Track-A1

Explore similar work

The Classics at SemEval-2026 Task 3: Combining Transformer Models and LLM-Generated Annotations for Dimensional Aspect-Based Sentiment Analysis
Jul 3, 2026Rafif Alshawi, Amit Raj, Aleksey Kudelya +1Sentiment Analysis Aspect-Based Sentiment Analysis
HSE University
ICT-NLP at SemEval-2026 Task 3: Less Is More -- Multilingual Encoder with Joint Training and Adaptive Ensemble for Dimensional Aspect Sentiment Regression
May 11, 2026Liyuan Huang, Jiawei He, Wutao Shen +2Aspect-Based Sentiment Analysis
1State Key Laboratory of AI Safety · Institute of Computing Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences
AILS-NTUA at SemEval-2026 Task 3: Efficient Dimensional Aspect-Based Sentiment Analysis
Mar 5, 2026Stavros Gazetas, Giorgos Filandrianos, Maria Lymperaiou +3Aspect-Based Sentiment Analysis
School of Electrical and Computer Engineering, AILS Laboratory National Technical University of Athens

Jul 3, 2026J/K move · Enter open · S save

The Classics at SemEval-2026 Task 3: Combining Transformer Models and LLM-Generated Annotations for Dimensional Aspect-Based Sentiment Analysis

Rafif Alshawi, Amit Raj, Aleksey Kudelya, Alexander Shirnin

HSE University

Sentiment Analysis Aspect-Based Sentiment Analysis

Open paper AskPDF arXiv

This paper presents an approach to the SemEval-2026 Task 3: Dimensional Aspect-Based Sentiment Analysis. We investigate methods for moving beyond traditional categorical sentiment (e.g., positive or negative) to predict fine-grained, real-valued scores for sentiment "valence" (positivity) and "arousal" (intensity). We participate in two subtasks: predicting these scores for given aspects (Subtask 1) and extracting full sets of sentiment details, including aspects, categories, and opinions alongside their scores (Subtask 3). Our approach for the regression task involves a weighted ensemble of transformer-based encoder models. For the Russian language, we further enhance the input by using a large language model (LLM) to generate synthetic sentiment descriptions. For the extraction task, we fine-tune a decoder LLM to perform structured prediction, allowing the system to identify sentiment elements and estimate their numerical scores simultaneously.

Abstract

Explore similar work

The Classics at SemEval-2026 Task 3: Combining Transformer Models and LLM-Generated Annotations for Dimensional Aspect-Based Sentiment Analysis

ICT-NLP at SemEval-2026 Task 3: Less Is More -- Multilingual Encoder with Joint Training and Adaptive Ensemble for Dimensional Aspect Sentiment Regression

AILS-NTUA at SemEval-2026 Task 3: Efficient Dimensional Aspect-Based Sentiment Analysis