做情感分析或文本分类的团队，DABS 用单次编码解决了多方面的计算冗余问题，直接复用编码结果能省 60% 算力，建议试试这个轻量方案。
多方面情感分析（ATSA）中，现有模型要么为每个方面重新编码句子，要么静态使用深层表示，导致计算冗余和适应性不足。DABS 提出单次推理框架，仅对句子编码一次，构建可复用的深度排序基板，每个方面通过查询该基板选择性读取相关 token 和抽象层级，无需重新编码。在四个基准测试中，DABS 在保持竞争性能的同时，将端到端计算量减少高达 60%，尤其在否定和对比等复杂语言场景中优势明显。代码已开源。
Single-Pass, Depth-Selective Reading for Multi-Aspect Sentiment Analysis
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