AI Paper Atlas

Research map / Large language models / Agents, code and math

From Solver to Tutor: Evaluating the Pedagogical Intelligence of LLMs with KMP-Bench

Weikang Shi, Houxing Ren, Junting Pan, Aojun Zhou, Ke Wang, Zimu Lu, Yunqiao Yang, Yuxuan Hu, Linda Wei, Mingjie Zhan, Hongsheng Li

AAAI 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-DialogueOverall Acc: 23.6 -> 37.0Improved
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-DialogueOverall Judgement: 21.0 -> 40.4Improved
fine-tune on KMP-Pile instead of MathChat_sync dataQwen2.5-Math-7B · KMP-DialogueOverall Acc: 15.3 -> 37.0Improved
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-ECMR-Score: 3.4 -> 49.0Improved
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-Gen Follow-uppass rate: 48.6 -> 81.1Improved
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-Gen Similarpass rate: 52.8 -> 83.2Improved
fine-tune on KMP-Pile tutoring dialoguesQwen2.5-Math-7B-Inst · KMP-FQAaccuracy: 76.1 -> 77.5Improved
use general-purpose model instead of math-tuned modelQwen2.5-72B · KMP-DialogueOverall Acc: 30.8 -> 58.3Improved
use math-tuned model instead of general-purpose modelQwen2.5-72B · KMP-Gen Follow-uppass rate: 81.3 -> 86.3Mixed
Gemini-2.0-Flash evaluator vs manual grading (ablation)KMP-Dialogue evaluation · 300-instance subsetAlignment Rate: 89.8% average alignmentImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy