Sherpa-Qwen3-8B

Sherpa_DP from Sherpa: Teaching LLMs to Teach Adaptively: a LoRA adapter for Qwen/Qwen3-8B, trained with students of diverse learning preferences. Code: SALT-NLP/Sherpa.

Usage

hf download SALT-NLP/Sherpa-Qwen3-8B --local-dir sherpa-qwen3-8b
python -m sglang.launch_server --model-path Qwen/Qwen3-8B --enable-lora \
  --lora-paths sherpa-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]
SherpaNP	63.1	58.8	52.0	45.3	55.1	40.6	49.2	54.8	48.3	52.0	[49.7, 54.5]
SherpaDP	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
SherpaNP	92.1	21.7	72.7	32.8	75.9	65.3	82.5	64.2	81.7	73.4
SherpaDP	93.2	21.9	73.5	30.3	77.0	72.0	90.3	67.3	87.2	79.2

Human pairwise comparison with Qwen3-8B: 96 high school teachers chose which reply teaches better, 240 judgments per request type; the win rate excludes ties.

Student request	SherpaDP	Ties	Base	Win rate
Subgoal decomposition	167	19	54	75.6%
Contrastive comparison	201	5	34	85.5%
Causal justification	185	10	45	80.4%
Error anticipation	175	11	54	76.4%
Overall	728	45	187	79.6%

Citation

@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},
}