The State-Prediction Separation Hypothesis

• URL: http://arxiv.org/abs/2607.01218v1
• Date: Wed, 01 Jul 2026 17:55:09 GMT
• Title: The State-Prediction Separation Hypothesis
• Abstract summary: Transformers use the same forward computation stream to both predict the next token and store useful state for future token predictions.<n>We design a Transformer variant that uses two computation streams to separate the two functions, and conduct pretraining experiments across various scales.
• Score: 24.312483891452185
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Transformers use the same forward computation stream to both predict the next token and store useful state for future token predictions. We formulate the \emph{state-prediction separation hypothesis}: disentangling the two roles yields better language modeling performance. We design a Transformer variant that uses two computation streams to separate the two functions, and conduct pretraining experiments across various scales. Our experiments show that state-prediction separation consistently offers better data and compute efficiencies, improving validation loss and outperforming standard Transformers by 2--3 percentage points on average on downstream tasks. We also conduct extensive empirical analysis that rules out potential confounders and demonstrates the fundamental difference in the gradients our design entails.

Related papers

• Tracing distinguishability through transformer processing with stochastic LayerNorm [0.0]
  We give representations volume, turning similarity into statistical distinguishability.<n>We realize this idea in pretrained transformers through a light-touch modification to LayerNorm.<n>These results establish distinguishability as a functionally grounded lens on transformer computation.
  arXiv  Detail & Related papers  (2026-08-31T12:57:38Z)
• Transformer-like Inference from Optimal Control [0.5161531917413708]
  This paper derives, from first principles, inference architectures that solve the same prediction problem.<n>For two model classes, the prediction objective is reformulated as an optimal control problem.<n>Experiments reveal that when the embedding dimension is insufficient, the transformer implicitly exploits non-Markovian structure.
  arXiv  Detail & Related papers  (2026-05-15T04:42:19Z)
• Diffusion-Inspired Reconfiguration of Transformers for Uncertainty Calibration [52.017716672255524]
  Uncertainty calibration in pre-trained transformers is critical for their reliable deployment in risk-sensitive applications.<n>We propose a diffusion-inspired reconfiguration of transformers in which each feature transformation block is modeled as a probabilistic mapping.<n>Our method achieves superior calibration and predictive accuracy compared to existing uncertainty-aware transformers.
  arXiv  Detail & Related papers  (2026-02-09T17:24:47Z)
• From Shortcut to Induction Head: How Data Diversity Shapes Algorithm Selection in Transformers [67.02076505996284]
  We study how the choice of pretraining data distribution steers a shallow transformer toward one behavior or the other.<n>Our results shed light on the algorithmic biases of pretrained transformers and offer conceptual guidelines for data-driven control of their learned behaviors.
  arXiv  Detail & Related papers  (2025-12-21T08:10:26Z)
• Diffusion Bridge or Flow Matching? A Unifying Framework and Comparative Analysis [57.614436689939986]
  Diffusion Bridge and Flow Matching have both demonstrated compelling empirical performance in transformation between arbitrary distributions.<n>We recast their frameworks through the lens of Optimal Control and prove that the cost function of the Diffusion Bridge is lower.<n>To corroborate these theoretical claims, we propose a novel, powerful architecture for Diffusion Bridge built on a latent Transformer.
  arXiv  Detail & Related papers  (2025-09-29T09:45:22Z)
• Understanding Data Influence with Differential Approximation [63.817689230826595]
  We introduce a new formulation to approximate a sample's influence by accumulating the differences in influence between consecutive learning steps, which we term Diff-In.<n>By employing second-order approximations, we approximate these difference terms with high accuracy while eliminating the need for model convexity required by existing methods.<n>Our theoretical analysis demonstrates that Diff-In achieves significantly lower approximation error compared to existing influence estimators.
  arXiv  Detail & Related papers  (2025-08-20T11:59:32Z)
• Flow-based Conformal Prediction for Multi-dimensional Time Series [9.900139803164372]
  We propose a novel conformal prediction method to address two key challenges by integrating Transformer and Normalizing Flow.<n>The Transformer encodes the historical context of time series, and normalizing flow learns the transformation from the base distribution to the distribution of non-conformity scores conditioned on the encoded historical context.<n>We demonstrate that our proposed method achieves smaller prediction regions compared to the baselines while satisfying the desired coverage through comprehensive experiments using simulated and real-world time series datasets.
  arXiv  Detail & Related papers  (2025-02-08T22:04:05Z)
• Comateformer: Combined Attention Transformer for Semantic Sentence Matching [11.746010399185437]
  We propose a novel semantic sentence matching model named Combined Attention Network based on Transformer model (Comateformer)<n>In Comateformer model, we design a novel transformer-based quasi-attention mechanism with compositional properties.<n>Our proposed approach builds on the intuition of similarity and dissimilarity (negative affinity) when calculating dual affinity scores.
  arXiv  Detail & Related papers  (2024-12-10T06:18:07Z)
• Non-asymptotic Convergence of Training Transformers for Next-token Prediction [48.9399496805422]
  Transformers have achieved extraordinary success in modern machine learning due to their excellent ability to handle sequential data. This paper provides a fine-grained non-asymptotic analysis of the training dynamics of a one-layer transformer. We show that the trained transformer presents non-token prediction ability with dataset shift.
  arXiv  Detail & Related papers  (2024-09-25T20:22:06Z)
• Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis [63.66763657191476]
  We show that efficient numerical training and inference algorithms as low-rank computation have impressive performance for learning Transformer-based adaption. We analyze how magnitude-based models affect generalization while improving adaption. We conclude that proper magnitude-based has a slight on the testing performance.
  arXiv  Detail & Related papers  (2024-06-24T23:00:58Z)
• In-Context Convergence of Transformers [63.04956160537308]
  We study the learning dynamics of a one-layer transformer with softmax attention trained via gradient descent. For data with imbalanced features, we show that the learning dynamics take a stage-wise convergence process.
  arXiv  Detail & Related papers  (2023-10-08T17:55:33Z)
• Transformer Uncertainty Estimation with Hierarchical Stochastic Attention [8.95459272947319]
  We propose a novel way to enable transformers to have the capability of uncertainty estimation. This is achieved by learning a hierarchical self-attention that attends to values and a set of learnable centroids. We empirically evaluate our model on two text classification tasks with both in-domain (ID) and out-of-domain (OOD) datasets.
  arXiv  Detail & Related papers  (2021-12-27T16:43:31Z)

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.