---
base_model: Qwen/Qwen3-8B
library_name: peft
license: apache-2.0
pipeline_tag: text-generation
tags:
- lora
- tutoring
- education
- reinforcement-learning
---

# Sherpa-NP-Qwen3-8B

Sherpa<sub>NP</sub> from [*Sherpa: Teaching LLMs to Teach Adaptively*](https://arxiv.org/abs/2610.08778): a LoRA adapter for
[Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B), trained only with the student that
has no learning preference. Code: [SALT-NLP/Sherpa](https://github.com/SALT-NLP/Sherpa).

## Usage

```bash
hf download SALT-NLP/Sherpa-NP-Qwen3-8B --local-dir sherpa-np-qwen3-8b
python -m sglang.launch_server --model-path Qwen/Qwen3-8B --enable-lora \
  --lora-paths sherpa-np-qwen3-8b --max-lora-rank 16
```

Use the prompts in `examples/sherpa/prompts.py` of the code repository, with
thinking off. Each reply is `<reasoning>...</reasoning>` followed by
`<output>...</output>` (shown to the student) or `<end></end>` (ends the dialogue).

## Results

From the paper. **Bold**: this model.

Student test accuracy after teaching (%), Qwen3-1.7B student, 528 MATH test problems
(36% unaided). In-distribution (ID) archetypes: Non. None, Att. Attempt, Sub. Subgoal,
Ctr. Contrast. Out-of-distribution (OOD): Cau. Causal, Stp. Step, Ver. Verification.
Ovl.: all seven.

| Teacher                 | Non.     | Att.     | Sub.     | Ctr.     | Cau.     | Stp.     | Ver.     | ID       | OOD      | Ovl.     | 95% CI           |
| ----------------------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | ---------------- |
| Qwen3-8B                | 53.2     | 50.7     | 48.4     | 44.1     | 54.4     | 47.8     | 41.5     | 49.1     | 47.9     | 48.6     | [46.1, 51.0]     |
| PedagogicalRL           | 63.0     | 59.4     | 54.7     | 44.4     | 59.0     | 44.3     | 43.8     | 55.4     | 49.1     | 52.7     | [50.3, 55.1]     |
| **Sherpa<sub>NP</sub>** | **63.1** | **58.8** | **52.0** | **45.3** | **55.1** | **40.6** | **49.2** | **54.8** | **48.3** | **52.0** | **[49.7, 54.5]** |
| Sherpa<sub>DP</sub>     | 76.2     | 78.2     | 72.7     | 60.7     | 77.1     | 51.8     | 66.4     | 72.0     | 65.1     | 69.0     | [66.8, 71.2]     |

MathTutorBench (%). PS problem solving, SQ Socratic questioning, SC solution
correctness, ML mistake location, MC mistake correction, Scaf. scaffolding, IF
instruction following, -H hard. Ped. Avg.: mean of Scaf., IF, Scaf.-H and IF-H.

| Teacher                 | PS       | SQ       | SC       | ML       | MC       | Scaf.    | IF       | Scaf.-H  | IF-H     | Ped. Avg. |
| ----------------------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | --------- |
| Qwen3-8B                | 92.1     | 25.3     | 70.6     | 37.1     | 74.8     | 33.1     | 74.0     | 32.4     | 70.6     | 52.5      |
| PedagogicalRL           | 93.0     | 21.9     | 70.9     | 18.7     | 73.9     | 52.8     | 76.6     | 43.4     | 80.1     | 63.2      |
| **Sherpa<sub>NP</sub>** | **92.1** | **21.7** | **72.7** | **32.8** | **75.9** | **65.3** | **82.5** | **64.2** | **81.7** | **73.4**  |
| Sherpa<sub>DP</sub>     | 93.2     | 21.9     | 73.5     | 30.3     | 77.0     | 72.0     | 90.3     | 67.3     | 87.2     | 79.2      |

## Citation

```bibtex
@misc{xu2026sherpateachingllmsteach,
      title={Sherpa: Teaching LLMs to Teach Adaptively},
      author={Weixian Xu and Yanzhe Zhang and Zora Zhiruo Wang and Changyu Chen and Diyi Yang},
      year={2026},
      eprint={2610.08778},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2610.08778},
}
```