## ðŸŽ¯DART-Math

Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving @ https://github.com/hkust-nlp/dart-math

- Paper â€¢ 2407.13690 â€¢ Published â€¢ 1

#### hkust-nlp/dart-math-hard

Viewer â€¢ Updated â€¢ 585k â€¢ 710 â€¢ 5Note Performance produced by `DART-Math-Hard` is usually but not necessarily **slightly better (~1% absolutely)** than `DART-Math-Uniform`. `DART-Math-DSMath-7B` achieves 53.6% on MATH and 86.8% on GSM8K for in-domain benchmarks, and 40.7% on CollegeMath as a representative out-of-domain benchmark.

#### hkust-nlp/dart-math-dsmath-7b-prop2diff

Text Generation â€¢ Updated â€¢ 64 â€¢ 1Note All the `DART-Math` models use the Alpaca prompt template: ``` Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n###Instruction:\n{query}\n\n### Response:\n ```

#### hkust-nlp/dart-math-llama3-8b-prop2diff

Text Generation â€¢ Updated â€¢ 14#### hkust-nlp/dart-math-mistral-7b-prop2diff

Text Generation â€¢ Updated â€¢ 13#### hkust-nlp/dart-math-llama3-70b-prop2diff

Text Generation â€¢ Updated â€¢ 9#### hkust-nlp/dart-math-uniform

Viewer â€¢ Updated â€¢ 591k â€¢ 119 â€¢ 4#### hkust-nlp/dart-math-dsmath-7b-uniform

Text Generation â€¢ Updated â€¢ 15#### hkust-nlp/dart-math-llama3-8b-uniform

Text Generation â€¢ Updated â€¢ 10#### hkust-nlp/dart-math-mistral-7b-uniform

Text Generation â€¢ Updated â€¢ 10#### hkust-nlp/dart-math-llama3-70b-uniform

Text Generation â€¢ Updated â€¢ 9

#### hkust-nlp/dart-math-pool-math

Viewer â€¢ Updated â€¢ 1.62M â€¢ 22 â€¢ 1Note The data pool synthesized from the query set of the **MATH** training set, containing **all answer-correct samples** and other metadata produced during the work. `DART-Math-*` datasets are extracted from `dart-math-pool-*` data pools.

#### hkust-nlp/dart-math-pool-math-query-info

Viewer â€¢ Updated â€¢ 7.5k â€¢ 4Note The **synthesis information of queries** from the **MATH** training set, such as the numbers of raw/correct samples of each synthesis job. Usually used with `dart-math-pool-math`.

#### hkust-nlp/dart-math-pool-gsm8k

Preview â€¢ Updated â€¢ 7Note Similar to `dart-math-pool-math`.

#### hkust-nlp/dart-math-pool-gsm8k-query-info

Viewer â€¢ Updated â€¢ 7.47k â€¢ 9 â€¢ 1Note Similar to `dart-math-pool-math-query-info`.

#### hkust-nlp/vrt-baseline

Viewer â€¢ Updated â€¢ 591k â€¢ 7Note The VRT baseline dataset used to train baseline models `*-VRT` in Table 2 of the paper.

#### hkust-nlp/gsm8k-fix

Viewer â€¢ Updated â€¢ 7.47k â€¢ 164