Research map / Large language models / Retrieval and knowledge
Unit Dependency Graph and Its Application to Arithmetic Word Problem Solving
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add unit dependency graph reasoning to LCA++ | Arithmetic word problem solver · AllArith | accuracy: 79.4 -> 81.7 | Helped |
| Add unit dependency graph reasoning to LCA++ | Arithmetic word problem solver · AllArithLex | accuracy: 63.6 -> 68.9 | Helped |
| Add unit dependency graph reasoning to LCA++ | Arithmetic word problem solver · AllArithTmpl | accuracy: 74.7 -> 79.5 | Helped |
| Add vertex information to the solver (ablation) | UnitDep · AllArith | accuracy: 80.3 -> 81.7 | Helped |
| Add vertex information to the solver (ablation) | UnitDep · AllArithLex | accuracy: 67.2 -> 68.9 | Helped |
| Add vertex information to the solver (ablation) | UnitDep · AllArithTmpl | accuracy: 77.1 -> 79.5 | Helped |
| Add edge information to the solver (ablation) | UnitDep · AllArith | accuracy: 79.9 -> 81.7 | Helped |
| Add edge information to the solver (ablation) | UnitDep · AllArithLex | accuracy: 64.1 -> 68.9 | Helped |
| Add edge information to the solver (ablation) | UnitDep · AllArithTmpl | accuracy: 75.7 -> 79.5 | Helped |
| Add graph-structure constraints (ablation) | Decompose UDG predictor · AllArith | accuracy: 70.9 -> 73.6 | Helped |
| Add graph-structure constraints (ablation) | Decompose UDG predictor · AllArithLex | accuracy: 62.9 -> 67.7 | Helped |
| Add graph-structure constraints (ablation) | Decompose UDG predictor · AllArithTmpl | accuracy: 65.5 -> 68.7 | Helped |
Closest papers by meaning
- Can LLMs Solve Longer Math Word Problems Better? (ICLR 2025 · similarity 0.86)
- A Compact Model for Mathematics Problem Representations Distilled from BERT (AAAI 2025 · similarity 0.85)
- MathPile: A Billion-Token-Scale Pretraining Corpus for Math (NeurIPS 2024 · similarity 0.84)
- Template-Driven LLM-Paraphrased Framework for Tabular Math Word Problem Generation (AAAI 2025 · similarity 0.84)
- Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages (ICLR 2025 · similarity 0.84)
- OMEGA: Can LLMs Reason Outside the Box in Math? Evaluating Exploratory, Compositional, and Transformative Generalization (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.