Research map / Graph learning / Graph self-supervised learning
SMART: A Situation Model for Algebra Story Problems via Attributed Grammar
Yining Hong, Qing Li, Ran Gong, Daniel Ciao, Siyuan Huang, Song-Chun Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k test (overall) | answer accuracy: 76.8 -> 79.5 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k test (Motion) | answer accuracy: 76.9 -> 79.8 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k test (Task) | answer accuracy: 79.0 -> 79.0 | No clear effect |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k test (Relation) | answer accuracy: 76.0 -> 77.9 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k test (Price) | answer accuracy: 83.6 -> 81.8 | Hurt |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k OOD (overall) | answer accuracy: 45.8 -> 63.2 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k OOD (Motion) | answer accuracy: 44.5 -> 65.0 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k OOD (Task) | answer accuracy: 47.4 -> 64.8 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k OOD (Relation) | answer accuracy: 55.1 -> 62.9 | Improved |
| Use SMART instead of the strongest reported baseline | algebra story problem solving · ASP6.6k OOD (Price) | answer accuracy: 45.3 -> 60.8 | Improved |
| Add relation reasoning to NER-only parsing (ablation) | SMART · ASP6.6k test | answer accuracy: around 20% | Improved |
| Add Seq2Seq relation detection to the initial relation extraction module (ablation) | SMART · ASP6.6k test | answer accuracy | Improved |
Closest papers by meaning
- Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages (ICLR 2025 · similarity 0.86)
- Neuro-Symbolic Data Generation for Math Reasoning (NeurIPS 2024 · similarity 0.86)
- MACM: Utilizing a Multi-Agent System for Condition Mining in Solving Complex Mathematical Problems (NeurIPS 2024 · similarity 0.86)
- Evaluating Large Vision-and-Language Models on Children's Mathematical Olympiads (NeurIPS 2024 · similarity 0.86)
- OMEGA: Can LLMs Reason Outside the Box in Math? Evaluating Exploratory, Compositional, and Transformative Generalization (NeurIPS 2025 · similarity 0.86)
- Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Process (ICLR 2025 · similarity 0.85)
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.
