Bayesian Repetition Penalty: A Principled Adjacent-Conditional Framework for Reversing Attention Collapse in Autoregressive Language Models

• URL: http://arxiv.org/abs/2607.22694v1
• Date: Fri, 17 Jul 2026 17:01:57 GMT
• Title: Bayesian Repetition Penalty: A Principled Adjacent-Conditional Framework for Reversing Attention Collapse in Autoregressive Language Models
• Abstract summary: We present a principled framework that penalises or compensates anomalous confidence arising from collapsed generation patterns.<n>The resulting self-normalising penalty ratio $R=f(m,n,p)/f(np,n,p)$ requires no ad hoc standardisation and admits a closed-form logit offset with zero approximation error.
• Score: 9.240704665480939
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Attention collapse in autoregressive language models -- manifested as repetitive token loops where the model becomes trapped in self-reinforcing attractors -- is a persistent pathology that existing decoding-time heuristics fail to address at its root cause. We present a principled framework that penalises or compensates anomalous confidence arising from collapsed generation patterns, by comparing a token's observed frequency against its corpus prior through an adjacent-conditional probability construction. The resulting self-normalising penalty ratio $R=f(m,n,p)/f(np,n,p)$ requires no ad hoc standardisation and admits a closed-form logit offset with zero approximation error. The correction is isolated from the loss gradient and accumulated into a frozen output-layer bias via exponential moving average, enabling deployment as a repair mechanism for models that have already collapsed without requiring intrusive modifications to standard training pipelines. Experimental validation on a 1.5B-parameter model demonstrates that the frozen-bias mechanism can rescue a model already trapped in a collapsed attractor, reducing 2-gram repetition from 0.073 to near 0 while preserving generation quality.

Related papers

• Koopman-Based Robust Model Predictive Control for Nonlinear Systems with Stochastic Intermittent Measurements [11.153670606528545]
  Intermittent state measurements pose fundamental challenges to model predictive control of constrained nonlinear systems.<n>This paper develops a Koopman-based MPC framework with probabilistically truncated soft constraints.
  arXiv  Detail & Related papers  (2026-09-02T04:08:49Z)
• Hacking Generative Perplexity: Why Unconditional Text Evaluation Needs Distributional Metrics [49.443264461057645]
  Diffusion and continuous flow-based language models have emerged as the leading non-autoregressive alternatives to language modeling.<n>Progress in both paradigms is overwhelmingly tracked by generative perplexity (gen-PPL): the per-token negative log-likelihood of samples under a frozen autoregressive (AR) scorer.<n>We argue that this metric is unsound. By construction, gen-PPL measures only predictability under the scoring AR, not grammaticality or semantic coherence.
  arXiv  Detail & Related papers  (2026-06-07T02:35:56Z)
• Joint Model and Data Sparsification via the Marginal Likelihood [53.29070892356214]
  We propose jointly learning individual feature and sample relevancies, enabling simultaneous model and data sparsification.<n>This symmetric pruning of model and data offers a natural extension that preserves conjugacy.<n> Empirical results across diverse regression tasks affirm that a joint ARD approach consistently yields both sparse and robust prediction models.
  arXiv  Detail & Related papers  (2026-05-28T13:26:53Z)
• The Need for an External Observer Formalizing the Sufficiency Gap: A Mathematical Extension of Mixture Identifiability and Contextual Grounding in Sequence Models [0.0]
  We construct a binary mixed-regime process with one deterministic textual regime and one random regime governed by an unobserved latent state.<n>The resulting entropy difference is not an ordinary optimization error; it is a sufficiency gap caused by marginalization over an unobserved state.<n>A corrective signal reverses the posterior odds induced by the textual history exactly when its fidelity exceeds the text-only posterior weight assigned to the misleading regime.
  arXiv  Detail & Related papers  (2026-05-26T08:53:11Z)
• Gradients Must Earn Their Influence: Unifying SFT with Generalized Entropic Objectives [22.29000001610794]
  Standard negative log-likelihood for Supervised Fine-Tuning (SFT) applies uniform token-level weighting.<n>This rigidity creates a two-fold failure mode: (i) overemphasizing low-probability targets can amplify gradients on noisy supervision and disrupt robust priors, and (ii) uniform weighting provides weak sharpening when the model is already confident.<n>Existing methods fail to resolve the resulting plasticity--stability dilemma, often suppressing necessary learning signals alongside harmful ones.<n>We introduce Dynamic Entropy Fine-Tuning (DEFT), a parameter-free objective that modulates the
  arXiv  Detail & Related papers  (2026-02-11T22:56:43Z)
• Entropy-Reservoir Bregman Projection: An Information-Geometric Unification of Model Collapse [3.533187668612022]
  We present EntropyReser Bregman Projection- ERBP, an information-geometric framework that unifies these phenomena.<n>Our theory yields a necessary condition for collapse, (ii) a sufficient condition that guarantees a non-language entropy floor, and (iii) closed-form rates that depend on sample size.
  arXiv  Detail & Related papers  (2025-12-16T19:50:03Z)
• Closed-Loop Transformers: Autoregressive Modeling as Iterative Latent Equilibrium [0.6820746164515952]
  We introduce the closed-loop prediction principle, which requires that models iteratively refine latent representations until reaching a self-consistent equilibrium.<n>We instantiate this principle as Equilibrium Transformers, which augment standard transformer layers with an Equilibrium Refinement Module.<n>Preliminary experiments on the binary parity task demonstrate +3.28% average improvement on challenging sequences, with gains reaching +8.07% where standard transformers approach random performance.
  arXiv  Detail & Related papers  (2025-11-26T20:02:59Z)
• From Denoising to Refining: A Corrective Framework for Vision-Language Diffusion Model [72.73512218682187]
  We introduce ReDiff, a refining-enhanced diffusion framework that teaches the model to identify and correct its own errors.<n>Our approach features a two-stage training process: first, we instill a foundational revision capability by training the model to revise synthetic errors; second, we implement a novel online self-correction loop.<n>This mistake-driven learning endows the model with the crucial ability to revisit and refine its already generated output, effectively breaking the error cascade.
  arXiv  Detail & Related papers  (2025-10-22T06:58:55Z)
• One-for-More: Continual Diffusion Model for Anomaly Detection [63.50488826645681]
  Anomaly detection methods utilize diffusion models to generate or reconstruct normal samples when given arbitrary anomaly images.<n>Our study found that the diffusion model suffers from severe faithfulness hallucination'' and catastrophic forgetting''<n>We propose a continual diffusion model that uses gradient projection to achieve stable continual learning.
  arXiv  Detail & Related papers  (2025-02-27T07:47:27Z)
• A Pseudo-Semantic Loss for Autoregressive Models with Logical Constraints [87.08677547257733]
  Neuro-symbolic AI bridges the gap between purely symbolic and neural approaches to learning. We show how to maximize the likelihood of a symbolic constraint w.r.t the neural network's output distribution. We also evaluate our approach on Sudoku and shortest-path prediction cast as autoregressive generation.
  arXiv  Detail & Related papers  (2023-12-06T20:58:07Z)
• When Does Confidence-Based Cascade Deferral Suffice? [69.28314307469381]
  Cascades are a classical strategy to enable inference cost to vary adaptively across samples. A deferral rule determines whether to invoke the next classifier in the sequence, or to terminate prediction. Despite being oblivious to the structure of the cascade, confidence-based deferral often works remarkably well in practice.
  arXiv  Detail & Related papers  (2023-07-06T04:13:57Z)

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.