arXiv：cs.CL· Zhifan Sun, Sebastian Gombert, Fabian Zehner, Leon Camus, Longwei Cong, Hendrik Drachsler·· 3 小时前AI 评分33
RUSPAN：将评分标准片段视为标签表示的联合 LLM 编码短答题评分框架
Rubric Spans are Label Representations: Joint LLM Encoding for Short Answer Scoring
AI 导读
RUSPAN 是一个以评分标准为条件的自动短答题评分（ASAS）框架，将评分标准描述视为语义标签表示，把题目、学生答案与所有候选评分等级序列化为单条序列，在一次 LM 前向中基于评分标准片段与整序列表示对等级做 listwise 打分。
正文
Abstract:Automatic Short Answer Scoring (ASAS) requires models that can score student responses against question-specific criteria while remaining efficient and transferable across rubric sets. We propose RUSPAN, a rubric-conditioned ASAS framework that treats rubric descriptions as semantic label representations. RUSPAN serialises the question context, student answer, and all candidate rubric levels into a single sequence, then scores the levels listwise from the rubric-span and whole-sequence representations produced in a single LM pass. We further introduce RUSPAN-RIM, in which a Rubric-Independent Mask prevents rubric spans from attending to one another, making rubric representations depend only on the answer and question context and preventing overfitting to rubric patterns during training for zero-shot transfer. On six ASAS benchmarks spanning English, German, and Portuguese, RUSPAN improves mono-benchmark scoring over discriminative and generative baselines, while RIM with position reindexing delivers consistent and substantial gains on PT-ASAG, the held-out benchmark with the strongest combined language and rubric-structure shift.
| Comments: | EMNLP2026 Main |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.09660 [cs.CL] |
| (or arXiv:2610.09660v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09660 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhifan Sun [view email]
[v1]
Wed, 7 Oct 2026 08:29:04 UTC (157 KB)
来源：arXiv：cs.CL · arxiv.org