# LookAlike: Consistent Distractor Generation in Math MCQs

arXiv: https://arxiv.org/abs/2505.01903 · HTML: https://mlsanity.com/paper/2505.01903/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Process reward models for reasoning, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 29 of its 53 references that are in the map.
- First posted on arXiv 3 May 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Process reward models for reasoning](https://mlsanity.com/alignment/process-reward-models-for-reasoning/)

## Builds on (most cited first, 29 of 29)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models](https://mlsanity.com/paper/1610.02424/) (2016)
- [Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study](https://mlsanity.com/paper/2404.10719/) (2024)
- [Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive](https://mlsanity.com/paper/2402.13228/) (2024)
- [Large Language Models for Data Annotation and Synthesis: A Survey](https://mlsanity.com/paper/2402.13446/) (2024)
- [Disentangling Length from Quality in Direct Preference Optimization](https://mlsanity.com/paper/2403.19159/) (2024)
- [Benchmarking Complex Instruction-Following with Multiple Constraints Composition](https://mlsanity.com/paper/2407.03978/) (2024)
- [Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer](https://mlsanity.com/paper/2405.16436/) (2024)
- [Training LLM-based Tutors to Improve Student Learning Outcomes in Dialogues](https://mlsanity.com/paper/2503.06424/) (2025)
- [Multi-Task Inference: Can Large Language Models Follow Multiple Instructions at Once?](https://mlsanity.com/paper/2402.11597/) (2024)
- [Do LLMs estimate uncertainty well in instruction-following?](https://mlsanity.com/paper/2410.14582/) (2024)
- [Distractor Generation in Multiple-Choice Tasks: A Survey of Methods, Datasets, and Evaluation](https://mlsanity.com/paper/2402.01512/) (2024)
- [Knowledge-Driven Distractor Generation for Cloze-style Multiple Choice Questions](https://mlsanity.com/paper/2004.09853/) (2020)
- [3D-Properties: Identifying Challenges in DPO and Charting a Path Forward](https://mlsanity.com/paper/2406.07327/) (2024)
- [A BERT-based Distractor Generation Scheme with Multi-tasking and Negative Answer Training Strategies](https://mlsanity.com/paper/2010.05384/) (2020)
- [The SIFo Benchmark: Investigating the Sequential Instruction Following Ability of Large Language Models](https://mlsanity.com/paper/2406.19999/) (2024)
- [Improving Socratic Question Generation using Data Augmentation and Preference Optimization](https://mlsanity.com/paper/2403.00199/) (2024)
- [ProcBench: Benchmark for Multi-Step Reasoning and Following Procedure](https://mlsanity.com/paper/2410.03117/) (2024)
- [Learning Multi-Step Reasoning by Solving Arithmetic Tasks](https://mlsanity.com/paper/2306.01707/) (2023)
- [Improving Automated Distractor Generation for Math Multiple-choice Questions with Overgenerate-and-rank](https://mlsanity.com/paper/2405.05144/) (2024)
- [Enhancing Distractor Generation for Multiple-Choice Questions with Retrieval Augmented Pretraining and Knowledge Graph Integration](https://mlsanity.com/paper/2406.13578/) (2024)
- [MalAlgoQA: Pedagogical Evaluation of Counterfactual Reasoning in Large Language Models and Implications for AI in Education](https://mlsanity.com/paper/2407.00938/) (2024)
- [Generating Plausible Distractors for Multiple-Choice Questions via Student Choice Prediction](https://mlsanity.com/paper/2501.13125/) (2025)
- [SEE-DPO: Self Entropy Enhanced Direct Preference Optimization](https://mlsanity.com/paper/2411.04712/) (2024)
- [Does Multiple Choice Have a Future in the Age of Generative AI? A Posttest-only RCT](https://mlsanity.com/paper/2412.10267/) (2024)

## Built on by (most cited first, 7 of 7)

- [SMART: Simulated Students Aligned with Item Response Theory for Question Difficulty Prediction](https://mlsanity.com/paper/2507.05129/) (2025)
- [BenchMarker: An Education-Inspired Toolkit for Highlighting Flaws in Multiple-Choice Benchmarks](https://mlsanity.com/paper/2602.06221/) (2026)
- [CATTO: Balancing Preferences and Confidence in Language Models](https://mlsanity.com/paper/2601.23096/) (2026)
- [Tailoring Diagnostic Modeling to Individual Learners: Personalized Distractor Generation via MCTS-Guided Reasoning Reconstruction](https://mlsanity.com/paper/2508.11184/) (2025)
- [The Ever-Evolving Science Exam](https://mlsanity.com/paper/2507.16514/) (2025)
- [Simulating Students or Sycophantic Problem Solving? On Misconception Faithfulness of LLM Simulators](https://mlsanity.com/paper/2605.12748/) (2026)
- [Incentives Of EdTech: A Systematic Review Of EduNLP Research](https://mlsanity.com/paper/2606.13691/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly used to generate distractors for multiple-choice questions (MCQs), especially in domains like math education. However, existing approaches are limited in ensuring that the generated distractors are consistent with common student errors. We propose LookAlike, a method that improves error-distractor consistency via preference optimization. Our two main innovations are: (a) mining synthetic preference pairs from model inconsistencies, and (b) alternating supervised fine-tuning (SFT) with Direct Preference Optimization (DPO) to stabilize training. Unlike prior work that relies on heuristics or manually annotated preference data, LookAlike uses its own generation inconsistencies as dispreferred samples, thus enabling scalable and stable training. Evaluated on a real-world dataset of 1,400+ math MCQs, LookAlike achieves 51.6% accuracy in distractor generation and 57.2% in error generation under LLM-as-a-judge evaluation, outperforming an existing state-of-the-art method (45.6% / 47.7%). These improvements highlight the effectiveness of preference-based regularization and inconsistency mining for generating consistent math MCQ distractors at scale.