中文

Single-Pass, Depth-Selective Reading for Multi-Aspect Sentiment Analysis

计算与语言 2026-05-21 v1 人工智能

摘要

Aspect-Term Sentiment Analysis (ATSA) in multi-aspect sentences faces a fundamental tradeoff between efficiency and expressiveness. Existing models either re-encode the sentence for each aspect or rely on static use of deep representations, leading to redundant computation and limited adaptivity. We argue that Transformer depth is a costly, queryable resource, and propose DABS, a single-pass inference framework that encodes each sentence once to construct a reusable, depth-ordered substrate. Each aspect then queries this shared representation to selectively read relevant tokens and abstraction levels, without re-encoding. This decouples shared sentence encoding from lightweight, aspect-conditioned readout. Experiments on four ATSA benchmarks show that DABS achieves competitive performance while reducing end-to-end computation by up to 60% in multi-aspect settings (M >= 2). Further analyses indicate that adaptive depth querying is most beneficial for linguistically complex cases such as negation and contrast. Code is publicly available at https://github.com/panzhzh/acl-dabs

关键词

引用

@article{arxiv.2605.20998,
  title  = {Single-Pass, Depth-Selective Reading for Multi-Aspect Sentiment Analysis},
  author = {Yan Xia and Zhuangzhuang Pan and Amirrudin Kamsin and Chee Seng Chan},
  journal= {arXiv preprint arXiv:2605.20998},
  year   = {2026}
}

备注

Accepted at ACL2026 (main). Our solution (DABS) reads the sentence once, then lets each aspect selectively query the right tokens and Transformer depths, cutting redundant computation while preserving ATSA accuracy