Dependency reasoning improves aspect sentiment quadruple prediction accuracy
Dependency Reasoning for Aspect Sentiment Quadruple Prediction
Computation and LanguageMachine Learning
Summary
Figuring out detailed opinions about products or services is challenging because it requires understanding how different words relate, not just which words appear nearby. The authors present DRIFT, a method that helps large language models by feeding them the exact structure of sentences, so they can better link aspects (like 'food'), opinions (like 'delicious'), categories, and sentiments. This approach also improves how the system judges its own answers, finding more correct results than traditional exact-match checks. The method performs better than previous ones on standard restaurant review datasets.
What this means in practice
- •For customer review analysis teams: Identify detailed sentiment relationships in reviews with higher accuracy using syntax-based prompts in language models.
- •For social media monitoring teams: Detect nuanced opinion structures in social media texts by incorporating dependency parsing into language model workflows.
Authors
Shima Esfandyar, Seywan Soleymanbaigi, Fardin Akhlaghian Tab, Fatemeh Daneshfar, Mourad Oussalah
Abstract
Aspect Sentiment Quadruple Prediction (ASQP) is fundamentally a relational reasoning problem: identifying aspect-opinion-category-sentiment quadruples requires modeling structured dependencies, not merely sequential co-occurrence. Existing LLM-based approaches incorporate syntactic information indirectly, via retrieval-based demonstration selection or multi-stage training, rather than as unified, single-pass generation-time context. We present DRIFT, which linearizes dependency parse trees and injects them directly into LLM prompts as explicit relational context, enabling the model to reason over aspect-opinion dependencies rather than infer them from token proximity alone. DRIFT further integrates retrieval-augmented in-context learning and a syntax-aware LLM-based adjudication mechanism that complements exact-match evaluation under span-boundary variation. On SemEval Rest15, Rest16, and ACOS, DRIFT achieves the strongest performance among compared generative ASQP methods, reaching 54.47% and 64.10% F1 on Rest15 and Rest16, with consistent gains across short- and medium-distance aspect-opinion configurations; adjudication further shows exact-match metrics systematically underestimate semantically correct predictions.