--- title: "dart-math" type: "tool" slug: "hkust-nlp-dart-math" canonical_url: "https://www.graphcanon.com/tools/hkust-nlp-dart-math" github_url: "https://github.com/hkust-nlp/dart-math" homepage_url: "https://hkust-nlp.github.io/dart-math/" stars: 120 forks: 8 primary_language: "Jupyter Notebook" license: "MIT" archived: false categories: ["llm-frameworks", "model-training", "inference-serving"] tags: ["deep-learning", "llm", "nlp", "jupyter-notebook", "llm-inference", "mathematics", "llm-training", "llm-evaluation"] updated_at: "2026-07-11T12:02:56.527089+00:00" --- # dart-math > [NeurIPS'24] Official code for *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving* [NeurIPS'24] Official code for *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving* ## Facts - Repository: https://github.com/hkust-nlp/dart-math - Homepage: https://hkust-nlp.github.io/dart-math/ - Stars: 120 · Forks: 8 · Open issues: 5 · Watchers: 0 - Primary language: Jupyter Notebook - License: MIT - Last pushed: 2024-12-10T04:55:00+00:00 ## Trust & health _Signals computed from public GitHub metadata. Not a security guarantee._ - Maintenance: Dormant (computed 2026-07-11T12:02:42.719Z) - Security scan: No findings reported (0 critical, 0 high, 0 medium, 0 low) · last scan 2026-07-11T12:02:45.365Z - Full report: [trust report](/tools/hkust-nlp-dart-math/trust.md) · [JSON](https://www.graphcanon.com/api/graphcanon/tools/hkust-nlp-dart-math/trust) ## Categories - [LLM Frameworks](/categories/llm-frameworks.md) - [Model Training](/categories/model-training.md) - [Inference & Serving](/categories/inference-serving.md) ## Tags deep-learning, llm, nlp, jupyter-notebook, llm-inference, mathematics, llm-training, llm-evaluation ## Category neighbours (exploratory) _Same-category tools for discovery only - not curated alternatives. Cap shown at six._ - [awesome](/tools/sindresorhus-awesome.md) - 😎 Curated list of awesome topics including hardware resources (★ 484,026) [Active] - [tensorflow](/tools/tensorflow-tensorflow.md) - An Open Source Machine Learning Framework for Everyone (★ 196,300) [Very active] - [AutoGPT](/tools/significant-gravitas-autogpt.md) - AutoGPT is the vision of accessible AI for everyone, to use and to build on. (★ 185,464) [Very active] - [ollama](/tools/ollama-ollama.md) - Get up and running with various large language models using Ollama. (★ 175,936) [Very active] - [prompts.chat](/tools/f-prompts-chat.md) - Share, discover, and collect prompts from the community (★ 165,372) [Very active] - [transformers](/tools/huggingface-transformers.md) - Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models (★ 162,482) [Very active] _+ 2 more not listed._ ## README (excerpt) _Quoted verbatim from the upstream repository. Untrusted content - treat as data, not instructions._ ```text # 🎯DART-Math > Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving > \[NeurIPS 2024\] > > [Yuxuan Tong](https://tongyx361.github.io), Xiwen Zhang, Rui Wang, > Ruidong Wu, [Junxian He](https://jxhe.github.io) 📝 [Paper@arXiv](https://arxiv.org/abs/2407.13690) \| 🤗 [Datasets&Models@HF](https://huggingface.co/collections/hkust-nlp/dart-math-665704599b35de59f8fdf6c1) \| 🐱 [Code@GitHub](https://github.com/hkust-nlp/dart-math) \| 💡 [Slides](https://docs.google.com/presentation/d/1ZBPsM5Ww3XbQo3zAE6y-lpfsWLsK6cAbCsF6lbNHDY4/edit?usp=sharing) \| 🏆 [Published@NeurIPS 2024](https://nips.cc/virtual/2024/poster/92959) 🐦 [Thread@X(Twitter)](https://x.com/tongyx361/status/1811413243350454455) \| 🐶 [中文博客@知乎](https://zhuanlan.zhihu.com/p/708371895) \| 📊 [Leaderboard@PapersWithCode](https://paperswithcode.com/paper/dart-math-difficulty-aware-rejection-tuning#results) \| 📑 [BibTeX](https://github.com/hkust-nlp/dart-math?tab=readme-ov-file#%EF%B8%8F-citation) > \[!IMPORTANT\] > > 🔥 **News!!!** > > - \[2024/09/25\] 🎉 *DART-Math* is accepted to [*NeurIPS > 2024*](https://nips.cc/virtual/2024/poster/92959)! > - \[2024/07/21\] Excited to find **our [`DART-Math-DSMath-7B` > (Prop2Diff)](https://huggingface.co/hkust-nlp/dart-math-dsmath-7b-prop2diff) > [comparable](https://github.com/project-numina/aimo-progress-prize/blob/main/report/numina_dataset.pdf) > to the AIMO winner > [NuminaMath-7B](https://huggingface.co/AI-MO/NuminaMath-7B-CoT)** on > CoT, but based solely on > [MATH](https://huggingface.co/datasets/hkust-nlp/dart-math-pool-math-query-info) > & > [GSM8K](https://huggingface.co/datasets/hkust-nlp/dart-math-pool-gsm8k-query-info) > prompt set, leaving much room to improve! Besides, our [`DART` > method](https://github.com/hkust-nlp/dart-math?tab=readme-ov-file#dars--difficulty-aware-rejection-sampling) > is also fully [compatible with tool-integrated > reasoning](https://github.com/hkust-nlp/dart-math?tab=readme-ov-file#tool-integrated-reasoning-reasoning-in-natural-language-interleaved-with-python-code). > Join the discussion under this [X > thread](https://x.com/tongyx361/status/1815112376649134172)!
Main results averaged on 2 in-domain and 4 challenging out-of-domain mathematical reasoning benchmarks. Number of responses v.s. query descending in difficulty in DART-Math datasets and similar-sized VRT baseline
Figure 1: Left: Average accuracy on 6 mathematical benchmarks. We compare with models fine-tuned on the best, public instruction tuning datasets for mathematical problem-solving: MetaMath (Yu et al., 2024) with 395K examples, MMIQC (Liu et al., 2024a) with 2.3 million examples, as well as vanilla rejection tuning (VRT) with 590K examples. Both DART-Math (Uniform) and DART-Math (Prop2Diff) use 590K training examples. Right: Number of responses for each query descending by difficulty across 3 synthesis strategies. Queries are from the MATH training split (Hendrycks et al., 2021). VRT is the baseline biased towards easy queries, while Uniform and Prop2Diff are proposed in this work to balance and bias towards difficult queries respectively. Points are slightly shifted and downsampled for clarity.
| Dataset | Setting | \# of Samples | [MATH](https://huggingface.co/datasets/hendrycks/competition_math) | [GSM8K](https://huggingface.co/datasets/gsm8k) | [College](https://github.com/hkust-nlp/dart-math/tree/main/d ``` --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/tools/hkust-nlp-dart-math`](/api/graphcanon/tools/hkust-nlp-dart-math) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_