Verification of the Implicit World Model in a Generative Model via Adversarial Sequences

• URL: http://arxiv.org/abs/2602.05903v1
• Date: Thu, 05 Feb 2026 17:18:22 GMT
• Title: Verification of the Implicit World Model in a Generative Model via Adversarial Sequences
• Authors: András Balogh, Márk Jelasity,
• Abstract summary: We propose adversarial sequence generation for verifying the soundness of the sequence model.<n>Our adversaries generate valid sequences so as to force the sequence model to generate an invalid next move prediction.<n>We train models on random as well as high-quality chess games, using several training recipes.
• Score: 9.081141443436536
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Generative sequence models are typically trained on sample sequences from natural or formal languages. It is a crucial question whether -- or to what extent -- sample-based training is able to capture the true structure of these languages, often referred to as the ``world model''. Theoretical results indicate that we can hope for soundness at best, that is, generating valid sequences, but not necessarily all of them. However, it is still important to have practical tools that are able to verify whether a given sequence model is sound. In this study, we focus on chess, as it is a domain that provides enough complexity while having a simple rule-based world model. We propose adversarial sequence generation for verifying the soundness of the sequence model. Our adversaries generate valid sequences so as to force the sequence model to generate an invalid next move prediction. Apart from the falsification of soundness, this method is also suitable for a more fine-grained analysis of the failure modes and the effects of different choices during training. To demonstrate this, we propose a number of methods for adversarial sequence generation and evaluate the approach on a large set of chess models. We train models on random as well as high-quality chess games, using several training recipes. We find that none of the models are sound, but some training techniques and dataset choices are able to improve soundness remarkably. We also investigate the potential application of board state probes in both our training and attack methods. Our findings indicate that the extracted board states have no causal role in next token prediction in most of the models.

Related papers

• Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling [60.63703438729223]
  We show how different architectures and training methods affect model multi-step reasoning capabilities.<n>We confirm that increasing model depth plays a crucial role for sequential computations.
  arXiv  Detail & Related papers  (2025-08-22T18:57:08Z)
• Learning-Order Autoregressive Models with Application to Molecular Graph Generation [52.44913282062524]
  We introduce a variant of ARM that generates high-dimensional data using a probabilistic ordering that is sequentially inferred from data.<n>We demonstrate experimentally that our method can learn meaningful autoregressive orderings in image and graph generation.
  arXiv  Detail & Related papers  (2025-03-07T23:24:24Z)
• Gumbel Counterfactual Generation From Language Models [64.55296662926919]
  We show that counterfactual reasoning is conceptually distinct from interventions.<n>We propose a framework for generating true string counterfactuals.<n>We show that the approach produces meaningful counterfactuals while at the same time showing that commonly used intervention techniques have considerable undesired side effects.
  arXiv  Detail & Related papers  (2024-11-11T17:57:30Z)
• Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models [0.0]
  We train a GPT model on Othello games and find that the model learned an internal representation of the board state. We extend this work into the more complex domain of chess, training on real games and investigating our model's internal representations. Unlike Li et al.'s prior synthetic dataset approach, our analysis finds that the model also learns to estimate latent variables like player skill to better predict the next character.
  arXiv  Detail & Related papers  (2024-03-21T18:53:23Z)
• SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking [60.109453252858806]
  A maximum-likelihood (MLE) objective does not match a downstream use-case of autoregressively generating high-quality sequences. We formulate sequence generation as an imitation learning (IL) problem. This allows us to minimize a variety of divergences between the distribution of sequences generated by an autoregressive model and sequences from a dataset. Our resulting method, SequenceMatch, can be implemented without adversarial training or architectural changes.
  arXiv  Detail & Related papers  (2023-06-08T17:59:58Z)
• Learning to Jump: Thinning and Thickening Latent Counts for Generative Modeling [69.60713300418467]
  Learning to jump is a general recipe for generative modeling of various types of data. We demonstrate when learning to jump is expected to perform comparably to learning to denoise, and when it is expected to perform better.
  arXiv  Detail & Related papers  (2023-05-28T05:38:28Z)
• Training Trajectories of Language Models Across Scales [99.38721327771208]
  Scaling up language models has led to unprecedented performance gains. How do language models of different sizes learn during pre-training? Why do larger language models demonstrate more desirable behaviors?
  arXiv  Detail & Related papers  (2022-12-19T19:16:29Z)
• Robust Attack Graph Generation [11.419463747286716]
  We present a method to learn automaton models that are more robust to input modifications. It iteratively aligns sequences to a learned model, modifies the sequences to their aligned versions, and re-learns the model.
  arXiv  Detail & Related papers  (2022-06-15T19:26:39Z)
• Uncertainty Estimation for Language Reward Models [5.33024001730262]
  Language models can learn a range of capabilities from unsupervised training on text corpora. It is often easier for humans to choose between options than to provide labeled data, and prior work has achieved state-of-the-art performance by training a reward model from such preference comparisons. We seek to address these problems via uncertainty estimation, which can improve sample efficiency and robustness using active learning and risk-averse reinforcement learning.
  arXiv  Detail & Related papers  (2022-03-14T20:13:21Z)
• Thought Flow Nets: From Single Predictions to Trains of Model Thought [39.619001911390804]
  When humans solve complex problems, they rarely come up with a decision right-away. Instead, they start with an intuitive decision reflecting upon it, spot mistakes, resolve contradictions and jump between different hypotheses.
  arXiv  Detail & Related papers  (2021-07-26T13:56:37Z)
• Is BERT a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models' Transferability [74.11825654535895]
  We investigate whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. We find that even on non-text data, the models pre-trained on text converge faster than the randomly models.
  arXiv  Detail & Related papers  (2021-03-12T09:19:14Z)

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.