{"success":true,"data":{"author":"Macina, Daheim, Chowdhury, Sinha, Kapur, Gurevych, Sachan (ETH Zurich, TU Darmstadt)","bucketAccessType":"public","bucketName":"eth-nlped/mathdial","bucketPrefix":"","category":"Mathematics","cloudProvider":"huggingface","createdAt":"2026-11-21T22:16:38.872Z","description":"MathDial is a dataset of one-to-one tutoring conversations grounded in grade-school math word problems from GSM8K. It was collected by pairing human teachers with a large language model that played a 7th-grade student holding a realistic misconception. There are 2,861 dialogues. Each records the problem text and correct answer, the student's incorrect solution and student profile, and the teacher's free-text description of the student's confusion. The teachers' own ratings are included too: whether the student solved the problem by the end (or the teacher had to reveal the answer), and how typical the confusion and interactions were on a 1 to 5 scale. Education researchers and learning scientists use it to study pedagogy in dialogue, for example when tutors scaffold and when they give away answers, and to train and evaluate tutoring models. Load it with Hugging Face datasets. MathDial is licensed CC BY 4.0, and the GSM8K problems are MIT. Cite Macina et al., MathDial, Findings of EMNLP 2023.","displayTitle":"MathDial: 2,861 Teacher-Student Tutoring Dialogues on Math Word Problems (ETH Zurich)","huggingFaceUrl":"https://huggingface.co/datasets/eth-nlped/mathdial","id":"2fce1cf7-79fa-5b86-a5a6-ba2d4fa7857c","license":"CC-BY-4.0","licenseEvidenceUrl":"https://github.com/eth-nlped/mathdial#license","objectCountEstimate":4,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2305.14536","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"eth-nlped/mathdial\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"eth-nlped/mathdial\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=eth-nlped, license:cc-by-4.0, size_categories:1K<n<10K, format:json, downloads=1236"],"format":"mixed","generatedAt":"2026-11-21T22:31:59.107Z","layout":{"example":"test.jsonl","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"{\"qid\": 6000025, \"scenario\": 1, \"question\": \"Julia was preparing for a dinner party at her house, where she intended to serve stew.  She noticed that she was out of plastic spoons, so she bought a new package of spoons.  Later, her husband also bought a package of 5 new spoons and gave them to Julia.  While Julia was making the stew, she used three of the spoons to sample her stew.  Later, when she went to set the table, she had a total of 12 spoons.  How many spoons were in the package that Julia bought?\",","encoding":"utf8","key":"test.jsonl","truncatedAt":512}],"sampleObjects":["test.jsonl","train.jsonl"],"sidecars":[{"bytes":"---\nlanguage:\n- en\nlicense: cc-by-4.0\ntags:\n- dialog\n- tutoring\n- math\n- gsm8k\n- conversation\n- education\ntask_categories:\n- text-generation\npretty_name: MathDial\nsize_categories:\n- 1K<n<10K\n---\n# Mathdial dataset\nhttps://arxiv.org/abs/2305.14536\n\nMathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problems.\n\nMathDial is grounded in math word problems as well as student confusions which provide a challenging testbed for creating faithful and equitable dialogue tutoring models able to reason over complex information. Current models achieve high accuracy in solving such problems but they fail in the task of teaching.\n\n## Data Structure\n- `qid` - unique identifier of the problem\n- `scenario` - order of the problem in the data collection, out of the 5 scenarios in a session\n- `question` - math problem text\n- `ground_truth` - correct answer to the problem\n- `student_incorrect_solution` - student incorrect solution to the problem caused by some confusion\n- `student_profile` - student profile based on general math problem solving student misconceptions\n- `teacher_described_confusion` - teacher annotated student confusion in free text\n- `self-correctness` - teacher annotated whether student solved the problem correctly by the end of the conversation\n    - options: `Yes`, `Yes, but I had to reveal the answer`, `No`\n- `self-typical-confusion` - teacher annotated whether student exhibited a typical 7th grade confusion, Likert scale 1 (unlikely) to 5 (very likely)\n- `self-typical-interactions` - teacher annotated whether student exhibited typical 7th grade interactions, Likert scale 1 (unlikely) to 5 (very likely)\n- `conversation` - conversation in a string format with `|EOM|` delimiter between Teacher and Student personas  `Persona: (dialog_act) text` e.g. `Teacher: (focus) What is the difference?|EOM|Student: I mean ...|EOM|Teacher:`\n\n\n---\nlicense: cc-by-4.0\n---","encoding":"utf8","key":"README.md","name":"README.md","size":1930,"truncated":false}],"sourceObjectKey":"test.jsonl","starterCell":"from datasets import load_dataset\nds = load_dataset(\"eth-nlped/mathdial\")\nprint(ds)","starterSamplePath":"test.jsonl","totalObjects":4},"summary":"MathDial is a dataset of one-to-one tutoring conversations grounded in grade-school math word problems from GSM8K.","title":"MathDial: 2,861 Teacher-Student Tutoring Dialogues on Math Word Problems (ETH Zurich)","updatedAt":"2026-11-22T00:24:53.181Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/eth-nlped/mathdial"}}