AGI Maze Prediction Datasets: A Compact Benchmark for Learning World Dynamics with Transformers

• URL: http://arxiv.org/abs/2609.02339v1
• Date: Wed, 02 Sep 2026 09:15:33 GMT
• Title: AGI Maze Prediction Datasets: A Compact Benchmark for Learning World Dynamics with Transformers
• Abstract summary: We introduce the AGI Maze Prediction datasets and Benchmark.<n> benchmark comprises per-step transition prediction, fixed-horizon state prediction, and sequential textual-observation prediction.<n>Results suggest structured, task-aligned working memory can be more useful than additional latent capacity alone.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: World modeling requires a predictive model to maintain and update an internal state adequate for reasoning about the consequences of actions. We introduce the AGI Maze Prediction Datasets and Benchmark, a lightweight controlled testbed for studying this capability in Transformers and other predictive models. Derived from procedurally generated, stateful grid worlds, the benchmark comprises per-step transition prediction, fixed-horizon state prediction, and sequential textual-observation prediction. Source-maze-disjoint training and validation splits, together with greedy exact-match evaluation, distinguish learning transferable action-conditioned dynamics from memorizing transitions in familiar layouts. We establish from-scratch byte-level Transformer baselines and compare them with two working-memory-augmented architectures. A generic auxiliary latent-memory Transformer can fit some training sets perfectly but does not consistently improve held-out performance. In contrast, a pseudo-video spatial-memory Transformer initializes a two-dimensional latent workspace from the input map and updates it from action history without receiving intermediate maps, positions, or state labels. Under the same data, objectives, and evaluation protocol, this model reaches perfect validation accuracy on selected fixed-horizon tasks where the byte and unstructured-memory baselines do not, and substantially improves sequential text-trace prediction. These results suggest that structured, task-aligned working memory can be more useful than additional latent capacity alone. More broadly, we argue that language grounding is mediated by persistent data structures and computations over them; the benchmark offers a compact setting for testing architectures that couple textual interfaces to learned structured state.

Related papers

• Parameter-Efficient Adaptation of Pretrained Language Models for Time-Series Forecasting [2.141079906482723]
  We project fixed-length time-series patches directly into the embedding space of a pretrained GPT-2 backbone.<n>We analyze the effects of (i)representation strategy (continuous embeddings versus textual serialisation), (ii)adaptation regime (frozen backbone versus partial or full fine-tuning), and (iv)input context length.
  arXiv  Detail & Related papers  (2026-09-14T10:30:59Z)
• ConfAL-WM: Confidence-Guided Active Learning for Action-Conditioned World Models [33.57978728341114]
  This paper presents ConfAL-WM, a confidence-guided active learning framework for post-training embodied world models.<n>Built upon EVAC, we attach a lightweight confidence to probe UNet decoder features and predict dense confidence maps in the latent space.<n>These maps are aggregated into task-, frame-, patch-level scores, enabling both efficient data selection and localized training enhancement.
  arXiv  Detail & Related papers  (2026-08-26T09:29:08Z)
• URDF-Anything: Constructing Articulated Objects with 3D Multimodal Language Model [76.08429266631823]
  We propose an end-to-end automatic reconstruction framework based on a 3D multimodal large language model (MLLM)<n>URDF-Anything utilizes an autoregressive prediction framework based on point-cloud and text multimodal input to jointly optimize geometric segmentation and kinematic parameter prediction.<n> Experiments on both simulated and real-world datasets demonstrate that our method significantly outperforms existing approaches.
  arXiv  Detail & Related papers  (2025-11-02T13:45:51Z)
• Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning [53.72709564555407]
  Memo is a transformer-based architecture and training recipe for reinforcement learning.<n>It incorporates the creation and retrieval of memory by interleaving periodic summarization tokens with the inputs of a model during training.<n>We demonstrate Memo's effectiveness on a gridworld meta-RL benchmark and a multi-object navigation task in photo-realistic indoor settings.
  arXiv  Detail & Related papers  (2025-10-22T16:24:47Z)
• Detect Anything via Next Point Prediction [51.55967987350882]
  Rex- Omni is a 3B-scale MLLM that achieves state-of-the-art object perception performance.<n>On benchmarks like COCO and LVIS, Rex- Omni attains performance comparable to or exceeding regression-based models.
  arXiv  Detail & Related papers  (2025-10-14T17:59:54Z)
• From Editor to Dense Geometry Estimator [77.21804448599009]
  We introduce textbfFE2E, a framework that adapts an advanced editing model based on Diffusion Transformer (DiT) architecture for dense geometry prediction.<n>FE2E achieves over 35% performance gains on the ETH3D dataset and outperforms the DepthAnything series, which is trained on 100$times$ data.
  arXiv  Detail & Related papers  (2025-09-04T15:58:50Z)
• Echo State Transformer: Attention Over Finite Memories [2.118933003468525]
  We introduce Echo State Transformers (EST), a hybrid architecture that elegantly resolves the challenge of sequential data processing.<n>EST integrates the Transformer attention mechanisms with principles from Reservoir Computing to create a fixed-size window distributed memory system.<n>EST ranks first overall in two of five categories, outperforming strong state-of-the-art baselines on classification and anomaly detection tasks.
  arXiv  Detail & Related papers  (2025-06-25T09:56:25Z)
• Unified Human Localization and Trajectory Prediction with Monocular Vision [64.19384064365431]
  MonoTransmotion is a Transformer-based framework that uses only a monocular camera to jointly solve localization and prediction tasks.<n>We show that by jointly training both tasks with our unified framework, our method is more robust in real-world scenarios made of noisy inputs.
  arXiv  Detail & Related papers  (2025-03-05T14:18:39Z)
• ORIGAMI: A generative transformer architecture for predictions from semi-structured data [3.5639148953570836]
  ORIGAMI is a transformer-based architecture that processes nested key/value pairs.<n>By reformulating classification as next-token prediction, ORIGAMI naturally handles both single-label and multi-label tasks.
  arXiv  Detail & Related papers  (2024-12-23T07:21:17Z)
• Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors [44.5740422079]
  We show that pretraining with standard denoising objectives leads to dramatic gains across multiple architectures. In stark contrast to prior works, we find vanilla Transformers to match the performance of S4 on Long Range Arena when properly pretrained.
  arXiv  Detail & Related papers  (2023-10-04T17:17:06Z)
• A Memory Transformer Network for Incremental Learning [64.0410375349852]
  We study class-incremental learning, a training setup in which new classes of data are observed over time for the model to learn from. Despite the straightforward problem formulation, the naive application of classification models to class-incremental learning results in the "catastrophic forgetting" of previously seen classes. One of the most successful existing methods has been the use of a memory of exemplars, which overcomes the issue of catastrophic forgetting by saving a subset of past data into a memory bank and utilizing it to prevent forgetting when training future tasks.
  arXiv  Detail & Related papers  (2022-10-10T08:27:28Z)
• Memory Transformer [0.31406146587437894]
  Transformer-based models have achieved state-of-the-art results in many natural language processing tasks. Memory-augmented neural networks (MANNs) extend traditional neural architectures with general-purpose memory for representations. We evaluate these memory augmented Transformers and demonstrate that presence of memory positively correlates with the model performance.
  arXiv  Detail & Related papers  (2020-06-20T09:06:27Z)

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.