GRADE: Generalizable Reasoning-Aware Dialogue Evaluation for AI Tutors

• URL: http://arxiv.org/abs/2605.27866v2
• Date: Thu, 04 Jun 2026 01:23:34 GMT
• Title: GRADE: Generalizable Reasoning-Aware Dialogue Evaluation for AI Tutors
• Abstract summary: GRADE is a systematic study of open-source models for pedagogical ability assessment in student-tutor dialogues.<n>We evaluate 120 configurations across five language models, zero-shot inference, LoRA fine-tuning, synthetic augmentation, CoT+Reasoning, and single-task versus multitask formulations.
• Score: 5.831342304669597
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Evaluating AI tutor responses requires more than factual correctness: tutors must identify mistakes, locate errors, provide guidance, and offer actionable next steps. We present GRADE, a systematic study of open-source models for pedagogical ability assessment in student-tutor dialogues. Building on the BEA 2025 TutorMind setting, we evaluate 120 configurations across five language models, zero-shot inference, LoRA fine-tuning, synthetic augmentation, CoT+Reasoning, and single-task versus multitask formulations. Gemma3-12B performs best for single-task evaluation, while Gemma3-27B in 8-bit precision is more reliable for multitask prediction. We find that augmentation helps models that struggle with the original data, verification adds limited gains despite higher cost, and CoT+Reasoning is more useful for synthetic data generation than direct classification. We further show that LoRA fine-tuning on structured classification objectives interferes with instruction-following behavior under thinking mode, redirecting generation away from the required evaluation format. Carbon analysis shows that model choice and reasoning mode substantially affect emissions. Overall, GRADE shows that carefully selected open-source LoRA pipelines can match or surpass proprietary and ensemble-based systems on key pedagogical dimensions, with code and data available at https://github.com/pvbgeek/GRADE.

Related papers

• CSTutorBench: Benchmarking Small Language Models as Tutors for Block-Based Programming [0.48342038441006796]
  Large language models are increasingly explored as AI tutors, yet deploying them in K-12 settings raises concerns around privacy, cost, and reliance on proprietary models.<n>Small language models (SLMs) offer a promising alternative, but selecting the right model for a specific educational context remains difficult.<n>We introduce CSTutorBench, a benchmark for evaluating language models as CS tutors in VEX VR, a block-based robotics environment.
  arXiv  Detail & Related papers  (2026-07-06T19:15:07Z)
• DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation [49.98710755440242]
  Distillation enables compact Vision-Language Models (VLMs) to obtain strong reasoning capabilities.<n>We reveal a critical inefficiency in this approach: up to 69% of the prompts in standard chart / document reasoning datasets are effectively zero-delta.<n>We propose a staged synthesis pipeline that repurposes existing datasets as seeds, actively targeting student failure modes to produce better prompts.
  arXiv  Detail & Related papers  (2026-05-15T02:04:12Z)
• SCRIBE: Structured Chain Reasoning for Interactive Behaviour Explanations using Tool Calling [9.113268651219187]
  SCRIBE is a framework for multi-hop, tool-augmented reasoning to generate valid responses to student questions about feedback reports.<n> Evaluation with a human-aligned GPT-Judge and a user study with 108 students shows that 8B-SCRIBE models achieve comparable or superior quality to much larger models.
  arXiv  Detail & Related papers  (2025-10-30T10:17:05Z)
• One Model to Critique Them All: Rewarding Agentic Tool-Use via Efficient Reasoning [54.580646706013965]
  Reward models (RMs) play a critical role in aligning large language models with human preferences.<n>We introduce ToolRM, a family of lightweight generative RMs tailored for general tool-use scenarios.<n>To build these models, we propose a novel pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling.
  arXiv  Detail & Related papers  (2025-10-30T06:08:27Z)
• TutorBench: A Benchmark To Assess Tutoring Capabilities Of Large Language Models [10.963195858672627]
  TutorBench is a dataset and evaluation benchmark designed to rigorously evaluate the core tutoring skills of large language models (LLMs)<n>Samples are drawn from three common tutoring tasks: (i) generating adaptive explanations tailored to a student's confusion, (ii) providing actionable feedback on a student's work, and (iii) promoting active learning through effective hint generation.<n>We evaluate 16 frontier LLMs on TutorBench and present a detailed analysis of their performance and behavior.
  arXiv  Detail & Related papers  (2025-10-03T01:41:09Z)
• MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors [82.91830877219822]
  We present MathTutorBench, an open-source benchmark for holistic tutoring model evaluation.<n>MathTutorBench contains datasets and metrics that broadly cover tutor abilities as defined by learning sciences research in dialog-based teaching.<n>We evaluate a wide set of closed- and open-weight models and find that subject expertise, indicated by solving ability, does not immediately translate to good teaching.
  arXiv  Detail & Related papers  (2025-02-26T08:43:47Z)
• Language Models are Few-Shot Graders [0.12289361708127876]
  We present an ASAG pipeline leveraging state-of-the-art LLMs.<n>We compare the grading performance of three OpenAI models: GPT-4, GPT-4o, and o1-preview.<n>Our findings indicate that providing graded examples enhances grading accuracy, with RAG-based selection outperforming random selection.
  arXiv  Detail & Related papers  (2025-02-18T23:38:21Z)
• ARES: Alternating Reinforcement Learning and Supervised Fine-Tuning for Enhanced Multi-Modal Chain-of-Thought Reasoning Through Diverse AI Feedback [13.154512864498912]
  We propose a two-stage algorithm ARES that Alternates REinforcement Learning (RL) and Supervised Fine-Tuning (SFT) First, we request the Teacher to score how much each sentence contributes to solving the problem in a Chain-of-Thought (CoT) Second, we ask the Teacher to correct the wrong reasoning after the RL stage. With the correction feedback, we stabilize the RL fine-tuned model through SFT.
  arXiv  Detail & Related papers  (2024-06-25T07:20:11Z)
• How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resources [117.6496550359768]
  This work explores recent advances in instruction-tuning language models on a range of open instruction-following datasets. We provide a large set of instruction-tuned models from 6.7B to 65B parameters in size, trained on 12 instruction datasets. We evaluate them on their factual knowledge, reasoning, multilinguality, coding, and open-ended instruction following abilities.
  arXiv  Detail & Related papers  (2023-06-07T19:59:23Z)
• Unifying Language Learning Paradigms [96.35981503087567]
  We present a unified framework for pre-training models that are universally effective across datasets and setups. We show how different pre-training objectives can be cast as one another and how interpolating between different objectives can be effective. Our model also achieve strong results at in-context learning, outperforming 175B GPT-3 on zero-shot SuperGLUE and tripling the performance of T5-XXL on one-shot summarization.
  arXiv  Detail & Related papers  (2022-05-10T19:32:20Z)
• Interpretable Learning-to-Rank with Generalized Additive Models [78.42800966500374]
  Interpretability of learning-to-rank models is a crucial yet relatively under-examined research area. Recent progress on interpretable ranking models largely focuses on generating post-hoc explanations for existing black-box ranking models. We lay the groundwork for intrinsically interpretable learning-to-rank by introducing generalized additive models (GAMs) into ranking tasks.
  arXiv  Detail & Related papers  (2020-05-06T01:51:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.