Can LLMs Estimate Student Struggles? Human-AI Difficulty Alignment with Proficiency Simulation for Item Difficulty Prediction

• URL: http://arxiv.org/abs/2512.18880v1
• Date: Sun, 21 Dec 2025 20:41:36 GMT
• Title: Can LLMs Estimate Student Struggles? Human-AI Difficulty Alignment with Proficiency Simulation for Item Difficulty Prediction
• Authors: Ming Li, Han Chen, Yunze Xiao, Jian Chen, Hong Jiao, Tianyi Zhou,
• Abstract summary: We present a large-scale empirical analysis of Human-AI Difficulty Alignment for over 20 models across diverse domains.<n>Our findings reveal a systematic misalignment where scaling up model size is not reliably helpful.<n>We observe that high performance often impedes accurate difficulty estimation, as models struggle to simulate the capability limitations of students.
• Score: 26.396483988509956
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate estimation of item (question or task) difficulty is critical for educational assessment but suffers from the cold start problem. While Large Language Models demonstrate superhuman problem-solving capabilities, it remains an open question whether they can perceive the cognitive struggles of human learners. In this work, we present a large-scale empirical analysis of Human-AI Difficulty Alignment for over 20 models across diverse domains such as medical knowledge and mathematical reasoning. Our findings reveal a systematic misalignment where scaling up model size is not reliably helpful; instead of aligning with humans, models converge toward a shared machine consensus. We observe that high performance often impedes accurate difficulty estimation, as models struggle to simulate the capability limitations of students even when being explicitly prompted to adopt specific proficiency levels. Furthermore, we identify a critical lack of introspection, as models fail to predict their own limitations. These results suggest that general problem-solving capability does not imply an understanding of human cognitive struggles, highlighting the challenge of using current models for automated difficulty prediction.

Related papers

• MentisOculi: Revealing the Limits of Reasoning with Mental Imagery [63.285794947638614]
  We develop MentisOculi, a suite of multi-step reasoning problems amenable to visual solution.<n> evaluating visual strategies ranging from latent tokens to explicit generated imagery, we find they generally fail to improve performance.<n>Our findings suggest that despite their inherent appeal, visual thoughts do not yet benefit model reasoning.
  arXiv  Detail & Related papers  (2026-02-02T18:49:06Z)
• Large Language Models as Students Who Think Aloud: Overly Coherent, Verbose, and Confident [0.8564319625930894]
  Large language models (LLMs) are increasingly embedded in AI-based tutoring systems. Can they faithfully model novice reasoning and metacognitive judgments?<n>We evaluate LLMs as novices using 630 think-aloud utterances from chemistry tutoring problems with problem-solving logs of student hint use, attempts, and problem context.<n>We compare LLM-generated reasoning to human learner utterances under minimal and extended contextual prompting, and assess the models' ability to predict step-level learner success.
  arXiv  Detail & Related papers  (2026-02-01T04:46:38Z)
• From Abstract to Contextual: What LLMs Still Cannot Do in Mathematics [79.81905350372067]
  We study gap through contextual mathematical reasoning.<n>We introduce ContextMATH, a benchmark that repurposes AIME and MATH-500 problems into two contextual settings.<n>Open-source models decline by 13 and 34 points on SG and CS, while proprietary models drop by 13 and 20.
  arXiv  Detail & Related papers  (2026-01-30T14:56:04Z)
• Four Quadrants of Difficulty: A Simple Categorisation and its Limits [4.304007567113229]
  We propose a four-quadrant categorisation of difficulty signals -- human vs. model and task-agnostic vs. task-dependent.<n>We find that task-agnostic features behave largely independently and that only task-dependent features align.<n>These findings challenge common Curriculum Learning intuitions and highlight the need for lightweight, task-dependent difficulty estimators.
  arXiv  Detail & Related papers  (2026-01-04T11:31:51Z)
• Vibe Reasoning: Eliciting Frontier AI Mathematical Capabilities -- A Case Study on IMO 2025 Problem 6 [28.84243696489176]
  We introduce Vibe Reasoning, a human-AI collaborative paradigm for solving complex mathematical problems.<n>We demonstrate this paradigm through IMO 2025 Problem 6, a optimization problem where autonomous AI systems publicly reported failures.
  arXiv  Detail & Related papers  (2025-12-22T11:30:19Z)
• THOUGHTTERMINATOR: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models [65.39456695678713]
  We introduce approximate measures of problem-level difficulty and demonstrate that a clear relationship between problem difficulty and optimal token spend exists.<n>We find that in general, reasoning models are poorly calibrated, particularly on easy problems.<n>We introduce THOUGHTTERMINATOR, a training-free black box decoding technique that significantly improves reasoning model calibration.
  arXiv  Detail & Related papers  (2025-04-17T22:16:30Z)
• Exploring the Potential of Large Language Models for Estimating the Reading Comprehension Question Difficulty [2.335292678914151]
  This study investigates the effectiveness of Large Language Models (LLMs) in estimating the difficulty of reading comprehension questions.<n>We use OpenAI's GPT-4o and o1, in estimating the difficulty of reading comprehension questions using the Study Aid and Reading Assessment (SARA) dataset.<n>The results indicate that while the models yield difficulty estimates that align meaningfully with derived IRT parameters, there are notable differences in their sensitivity to extreme item characteristics.
  arXiv  Detail & Related papers  (2025-02-25T02:28:48Z)
• Causality can systematically address the monsters under the bench(marks) [64.36592889550431]
  Benchmarks are plagued by various biases, artifacts, or leakage.<n>Models may behave unreliably due to poorly explored failure modes.<n> causality offers an ideal framework to systematically address these challenges.
  arXiv  Detail & Related papers  (2025-02-07T17:01:37Z)
• Are You Doubtful? Oh, It Might Be Difficult Then! Exploring the Use of Model Uncertainty for Question Difficulty Estimation [12.638577140117702]
  We show that uncertainty features contribute substantially to difficulty prediction, where difficulty is inversely proportional to the number of students who can correctly answer a question.<n>In addition to showing the value of our approach, we also observe that our model achieves state-of-the-art results on the USMLE and CMCQRD publicly available datasets.
  arXiv  Detail & Related papers  (2024-12-16T14:55:09Z)
• The Superalignment of Superhuman Intelligence with Large Language Models [63.96120398355404]
  We discuss the concept of superalignment from the learning perspective to answer this question.<n>We highlight some key research problems in superalignment, namely, weak-to-strong generalization, scalable oversight, and evaluation.<n>We present a conceptual framework for superalignment, which consists of three modules: an attacker which generates adversary queries trying to expose the weaknesses of a learner model; a learner which will refine itself by learning from scalable feedbacks generated by a critic model along with minimal human experts; and a critic which generates critics or explanations for a given query-response pair, with a target of improving the learner by criticizing.
  arXiv  Detail & Related papers  (2024-12-15T10:34:06Z)
• MacGyver: Are Large Language Models Creative Problem Solvers? [87.70522322728581]
  We explore the creative problem-solving capabilities of modern LLMs in a novel constrained setting.<n>We create MACGYVER, an automatically generated dataset consisting of over 1,600 real-world problems.<n>We present our collection to both LLMs and humans to compare and contrast their problem-solving abilities.
  arXiv  Detail & Related papers  (2023-11-16T08:52:27Z)
• The Generative AI Paradox: "What It Can Create, It May Not Understand" [81.89252713236746]
  Recent wave of generative AI has sparked excitement and concern over potentially superhuman levels of artificial intelligence. At the same time, models still show basic errors in understanding that would not be expected even in non-expert humans. This presents us with an apparent paradox: how do we reconcile seemingly superhuman capabilities with the persistence of errors that few humans would make?
  arXiv  Detail & Related papers  (2023-10-31T18:07:07Z)

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.