一次编码，按需读取，让情感分析更快更准。
Single-Pass, Depth-Selective Reading for Multi-Aspect Sentiment Analysis
- 单次编码句子，构建深度有序的共享表示
- 多方面分析时计算量减少60%，性能仍领先
- 擅长处理否定、对比等复杂语言现象
多方面情感分析在效率与表达力之间存在根本权衡。现有模型要么为每个方面重编码句子，要么依赖静态深层表征，导致冗余计算且适应性差。本文认为Transformer深度是高成本但可查询的资源，提出DABS框架：仅对句子进行一次编码，生成可复用的深度有序基底，各方面通过查询该共享表示，选择性读取相关词元与抽象层级，无需重编码。该方法将共享编码与轻量级、面向方面的读取解耦。在四个ATSA基准上实验表明，DABS在多方面场景（M ≥ 2）下端到端计算量最多降低60%，同时保持竞争力。进一步分析显示，自适应深度查询在否定、对比等语言复杂情况下收益最大。代码已公开于https://github.com/panzhzh/acl-dabs。
原文摘要 · Abstract (English)
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
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。