Research map / Vision-language and multimodal / Multimodal LLMs
OX-MABSR: A Benchmark for Open-domain Explainable Multimodal Aspect-Based Sentiment Reasoning
Xinjing Liu, ZiXin Xue, Pengyue Lin, Xinyu Tu, Siwei Xu, Ruifan Li
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Introduces OX-MABSR, a benchmark and task for open-domain explainable multimodal aspect-based sentiment reasoning covering expressive sentiment categories and implicit aspects.
Problem
Existing multimodal aspect-based sentiment analysis (MABSA) tasks usually use a fixed sentiment set and explicitly mentioned aspects, and rarely address expressive sentiment categories, implicit aspects, or explainability.
Approach
Defines a new open-domain explainable MABSA task that extracts aspects from text-image pairs (including implicit ones), assigns expressive sentiment categories, and requires explanations, released as a benchmark.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add images (multimodal input) to MABSR-LLM (ablation) | MABSR-LLM · OX-MABSR-Bench test, aspect-sentiment | F1_in: 31.15 -> 44.65 | Improved |
| add images (multimodal input) (ablation) | MABSR-LLM · OX-MABSR-Bench test, reasoning paths | SC: 52.78 -> 72.17 | Improved |
| add GRPO RL stage after SFT (ablation) | MABSR-LLM · OX-MABSR-Bench test, aspect-sentiment | F1_in: 38.61 -> 44.65 | Improved |
| add GRPO RL stage after SFT (ablation) | MABSR-LLM · OX-MABSR-Bench test, perceptual causes | SC: 62.87 -> 71.21 | Improved |
| add reasoning-path generation task to multi-task training (ablation) | MABSR-LLM · OX-MABSR-Bench test, reasoning paths | SC: 60.77 -> 72.17 | Improved |
| add reasoning-path generation task to multi-task training (ablation) | MABSR-LLM · OX-MABSR-Bench test, aspect-sentiment | F1_in: 41.72 -> 44.65 | Improved |
| use Qwen3 backbone instead of LLaMA3-8B or ChatGLM3-6B (ablation) | MABSR-LLM · OX-MABSR-Bench test | SC | Improved |
| set LoRA rank to 8 instead of 4 (ablation) | MABSR-LLM · OX-MABSR-Bench test | reasoning quality | Improved |
| raise LoRA rank from 8 to 16 (ablation) | MABSR-LLM · OX-MABSR-Bench test | reasoning quality | Hurt |
| train MABSR-LLM instead of prompting general MLLMs | InternVL3-8B baseline · OX-MABSR-Bench test, aspect-sentiment | F1_in: 15.08 -> 44.65 | Improved |
| implicit rather than explicit aspects | MABSR-LLM · OX-MABSR-Bench test, aspect-sentiment | F1 | Hurt |
| longer, harder explanations | MABSR-LLM · OX-MABSR-Bench test, explanations | SC | Hurt |
Experimental setup
Benchmarks: OX-MABSR
Builds on: MABSA
Closest papers by meaning
- DEQA: Descriptions Enhanced Question-Answering Framework for Multimodal Aspect-Based Sentiment Analysis (AAAI 2025 · similarity 0.90)
- Aspect Enhancement and Text Simplification in Multimodal Aspect-Based Sentiment Analysis for Multi-Aspect and Multi-Sentiment Scenarios (AAAI 2025 · similarity 0.90)
- A Novel Energy Based Model Mechanism for Multi-Modal Aspect-Based Sentiment Analysis (AAAI 2024 · similarity 0.88)
- You Only Read Once: Constituency-Oriented Relational Graph Convolutional Network for Multi-Aspect Multi-Sentiment Classification (AAAI 2024 · similarity 0.85)
- DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis (AAAI 2025 · similarity 0.85)
- SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Data (NeurIPS 2026 · similarity 0.85)
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.
