Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering

• URL: http://arxiv.org/abs/2605.27249v1
• Date: Tue, 26 May 2026 16:27:07 GMT
• Title: Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering
• Abstract summary: An ideal learning demonstration is a counterfactual version of the student work, an improved version that is still similar to their own.<n>Existing automated approaches for counterfactual text generation using Large Language Models (LLMs) result in domain-specific systems that are difficult to translate into practical applications.<n>We present the Gumbel Machine, a flexible, modular approach to generating counterfactuals that leverages LLM instruction-following capabilities while encouraging similarity to a reference factual text.
• Score: 41.340472784619756
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: An effective method of teaching across disciplines is to provide examples of high-quality work. However, an example may be significantly different from a student's current work, making it challenging for them to emulate. An ideal learning demonstration is a counterfactual version of the student work, an improved version that is still similar to their own. Existing automated approaches for counterfactual text generation using Large Language Models (LLMs) result in domain-specific systems that are difficult to translate into practical applications. We present the Gumbel Machine, a flexible, modular approach to generating counterfactuals that leverages LLM instruction-following capabilities while encouraging similarity to a reference factual text. Central to our approach is a novel, controlled decoding algorithm, $β$-Hindsight control, which uses latent randomness as a tunable similarity control mechanism during counterfactual generation. Experiments on datasets of student writing, scored on various criteria, demonstrate the effectiveness of our approach at generating counterfactuals both rubric-consistent and similar to a reference.

Related papers

• Teaching Language Models How to Code Like Learners: Conversational Serialization for Student Simulation [4.361813083295904]
  We propose a method for training open-weight artificial programming learners using authentic student process data.<n>Our approach serializes temporal log traces into a conversational format, representing each student's problem-solving process.<n>Student code submissions and environment feedback, such as test outcomes, grades, and error traces, form alternating conversational turns.
  arXiv  Detail & Related papers  (2026-04-12T16:38:42Z)
• GRAD: Generative Retrieval-Aligned Demonstration Sampler for Efficient Few-Shot Reasoning [6.371118178107715]
  Large Language Models (LLMs) achieve strong performance across diverse tasks.<n>Retrieval-Augmented Generation (RAG) enriches prompts with external information, but its reliance on static databases constrains adaptability.<n>We propose a Generative Retrieval-Aligned Demonstrator (GRAD) to generate input-specific concise demonstrations.
  arXiv  Detail & Related papers  (2025-10-01T17:52:41Z)
• ParaStudent: Generating and Evaluating Realistic Student Code by Teaching LLMs to Struggle [24.691302820912888]
  Large Language Models (LLMs) have shown strong performance on programming tasks, but can they generate student-like code like real students?<n>We present ParaStudent, a systematic study of LLM-based "student-like" code generation in an introductory programming course setting.
  arXiv  Detail & Related papers  (2025-07-16T23:12:14Z)
• Automated Generation of Commit Messages in Software Repositories [0.7366405857677226]
  Commit messages are crucial for documenting software changes, aiding in program comprehension and maintenance.<n>Our research presents an automated approach to generate commit messages using Machine Learning (ML) and Natural Language Processing (NLP)<n>We used the dataset of code changes and corresponding commit messages that was used by Liu et al.
  arXiv  Detail & Related papers  (2025-04-17T15:08:05Z)
• Enhancing Text Generation in Joint NLG/NLU Learning Through Curriculum Learning, Semi-Supervised Training, and Advanced Optimization Techniques [0.0]
  This research paper developed a novel approach to improve text generation in the context of joint Natural Language Generation (NLG) and Natural Language Understanding (NLU) learning.<n>The data is prepared by gathering and preprocessing annotated datasets, including cleaning, tokenization, stemming, and stop-word removal.<n> Transformer-based encoders and decoders, capturing long range dependencies and improving source-target sequence modelling.<n>Reinforcement learning with policy gradient techniques, semi-supervised training, improved attention mechanisms, and differentiable approximations are employed to fine-tune the models and handle complex linguistic tasks effectively.
  arXiv  Detail & Related papers  (2024-10-17T12:43:49Z)
• Aligning Language Models with Demonstrated Feedback [58.834937450242975]
  Demonstration ITerated Task Optimization (DITTO) directly aligns language model outputs to a user's demonstrated behaviors.<n>We evaluate DITTO's ability to learn fine-grained style and task alignment across domains such as news articles, emails, and blog posts.
  arXiv  Detail & Related papers  (2024-06-02T23:13:56Z)
• DiffGen: Robot Demonstration Generation via Differentiable Physics Simulation, Differentiable Rendering, and Vision-Language Model [72.66465487508556]
  DiffGen is a novel framework that integrates differentiable physics simulation, differentiable rendering, and a vision-language model. It can generate realistic robot demonstrations by minimizing the distance between the embedding of the language instruction and the embedding of the simulated observation. Experiments demonstrate that with DiffGen, we could efficiently and effectively generate robot data with minimal human effort or training time.
  arXiv  Detail & Related papers  (2024-05-12T15:38:17Z)
• Text2Data: Low-Resource Data Generation with Textual Control [100.5970757736845]
  Text2Data is a novel approach that utilizes unlabeled data to understand the underlying data distribution.<n>It undergoes finetuning via a novel constraint optimization-based learning objective that ensures controllability and effectively counteracts catastrophic forgetting.
  arXiv  Detail & Related papers  (2024-02-08T03:41:39Z)
• $\textit{latent}$-GLAT: Glancing at Latent Variables for Parallel Text Generation [65.29170569821093]
  parallel text generation has received widespread attention due to its success in generation efficiency. In this paper, we propose $textitlatent$-GLAT, which employs the discrete latent variables to capture word categorical information. Experiment results show that our method outperforms strong baselines without the help of an autoregressive model.
  arXiv  Detail & Related papers  (2022-04-05T07:34:12Z)
• Learning Multi-Objective Curricula for Deep Reinforcement Learning [55.27879754113767]
  Various automatic curriculum learning (ACL) methods have been proposed to improve the sample efficiency and final performance of deep reinforcement learning (DRL) In this paper, we propose a unified automatic curriculum learning framework to create multi-objective but coherent curricula. In addition to existing hand-designed curricula paradigms, we further design a flexible memory mechanism to learn an abstract curriculum.
  arXiv  Detail & Related papers  (2021-10-06T19:30:25Z)
• Unsupervised Text Generation by Learning from Search [86.51619839836331]
  TGLS is a novel framework to unsupervised Text Generation by Learning. We demonstrate the effectiveness of TGLS on two real-world natural language generation tasks, paraphrase generation and text formalization.
  arXiv  Detail & Related papers  (2020-07-09T04:34:48Z)

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.