MultiGranular Sentiment Intensity Augmentation with Ordinal Contrastive Learning for Enhanced Sentiment Classification
Bibliographic record
Abstract
Current sentiment analysis approaches rely on coarse-grained classification, failing to capture the nuanced intensity variations that characterize human emotional expression. We propose Multi-Granular Sentiment Intensity Augmentation (MGSIA), a systematic framework that generates five-level sentiment intensity spectra using Large Language Models (LLMs) and employs ordinal contrastive learning to preserve intensity relationships in embedding space.MGSIA combines three key innovations: (1) LLM-based intensity-aware data augmentation that generates semantically coherent sentiment variations, (2) ordinal contrastive learning that respects intensity hierarchies, and (3) a unified loss function integrating classification, ordinal consistency, and contrastive objectives.Comprehensive experiments on three benchmark datasets (SST-5, IMDB, and Yelp) demonstrate consistent improvements over state-of-the-art methods, with an average accuracy gain of 2.3% across all datasets. MGSIA achieves particularly strong performance on challenging neutral sentiment classification (+5.6% improvement) and fine-grained sentiment tasks (+3.1% on SST-5). Ablation studies confirm that each component contributes meaningfully, with the complete framework yielding a total 5.2% accuracy improvement over baseline approaches. Our approach addresses a gap in sentiment analysis by providing both robust classification performance and interpretable intensity-aware predictions for enhanced sentiment understanding.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".