Matrix-Decoupled Concentration for Autoregressive Sequences: Dimension-Free Guarantees for Sparse Long-Context Rewards

• URL: http://arxiv.org/abs/2605.06017v1
• Date: Thu, 07 May 2026 11:12:59 GMT
• Title: Matrix-Decoupled Concentration for Autoregressive Sequences: Dimension-Free Guarantees for Sparse Long-Context Rewards
• Abstract summary: We show how to establish tight concentration bounds for Sequence-level evaluations in autoregressive Large Language Models.<n>By preserving the coordinate-wise sparsity of rewards within a strictly causal framework, we provide a rigorous mathematical justification for the stability of long-context reasoning.
• Score: 0.36260136172126667
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Sequence-level evaluations in autoregressive Large Language Models (LLMs) rely on highly dependent token generation. Establishing tight concentration bounds for these processes remains a challenge due to two fundamental bottlenecks in existing frameworks: (i) classical inequalities typically separate dependency structures from target sensitivities, leading to a scalar collapse that inflates the variance proxy to a suboptimal $\mathcal{O}(N)$ for sparse terminal rewards; (ii) conversely, while certain spatial methods achieve tighter bounds, they lack the strictly causal filtration required by sequential generation, rendering them inapplicable to the autoregressive setting. To resolve both bottlenecks, we establish a sharp McDiarmid-type inequality for dependent sequences, governed strictly by the exact matrix-vector multiplication of the causal dependency resolvent and the target sensitivity vector. This Matrix-Decoupled Concentration (MDC) framework natively recovers optimal constants for Markov chains and exploits directed $d$-separation to yield order-optimal bounds for causal trees. Crucially, by exactly preserving the coordinate-wise sparsity of rewards within a strictly causal framework, MDC mathematically prevents scalar collapse, guaranteeing a dimension-free $\mathcal{O}(1)$ variance proxy and providing a rigorous mathematical justification for the stability of long-context reasoning.

Related papers

• C$^{2}$R: Cross-sample Consistency Regularization Mitigates Feature Splitting and Absorption in Sparse Autoencoders [56.67585330425431]
  Sparse Autoencoders (SAEs) are widely used to interpret large language models by decomposing activations into sparse, human-understandable features.<n>These issues stem from inconsistent latent assignment across samples.<n>We introduce C$2$R, which penalizes the co-activation of directionally similar latents.
  arXiv  Detail & Related papers  (2026-06-29T17:45:31Z)
• Expanding SPHERE-JEPA: A Family of Statistical Regularizers for the Hypersphere [11.087393998848254]
  In Self-Supervised Learning (SSL), preventing representation collapse by explicitly enforcing a uniform distribution on the unit hypersphere has proven to be effective.<n>Current frameworks typically rely on sliced statistical regularizers such as SIGReg (used in LeJEPA) and SUSReg (used in SPHEREJEPA)<n>This injects projection variance into the training gradients, destabilizing optimization, and hindering convergence.<n>In this work, we first show that analytically integrating out these random projections yields a deterministic Maximum Mean Discrepancy (MMD), bypassing the variance of sliced methods.
  arXiv  Detail & Related papers  (2026-06-16T07:10:05Z)
• PAC-Bayesian Adversarially Robust Generalization for Message Passing Graph Neural Networks: A Sensitivity Analysis [80.64607823130437]
  We extend PAC-Bayesian framework from deep neural networks to message passing GNNs.<n>We derive a tighter robust generalization bound in the adversarial setting.
  arXiv  Detail & Related papers  (2026-06-04T15:31:21Z)
• Model Compression with Exact Budget Constraints via Riemannian Manifolds [39.54576236079211]
  Assigning one of K options to each of N groups under a total cost budget is a recurring problem in efficient AI.<n>We present a new approach by showing that, under softmax relaxation, the budget constraint defines a smooth Riemannian manifold in logit space with unusually simple geometry.<n>Building on these properties, we propose Riemannian Constrained Optimization (RCO), which augments a standard Adam step with tangent projection, binary-search retraction, and momentum transport.
  arXiv  Detail & Related papers  (2026-05-01T13:30:23Z)
• Adaptive Threshold-Driven Continuous Greedy Method for Scalable Submodular Optimization [0.0]
  Submodular under matroid constraints is a fundamental problem in optimization with applications in sensing, data-marization, active learning, and resource allocation.<n>We propose textitATCG (underlineAdaptive underlineThresholded underlineThresholded underlineGreedy), expanding each agent's active set only when current candidates fail to capture sufficient marginal gain.<n>Experiments on a classbalanced prototype selection problem over a CIFAR-10 animal dataset show that textit
  arXiv  Detail & Related papers  (2026-04-03T19:32:39Z)
• Regularized Online RLHF with Generalized Bilinear Preferences [68.44113000390544]
  We consider the problem of contextual online RLHF with general preferences.<n>We adopt the Generalized Bilinear Preference Model to capture preferences via low-rank, skew-symmetric matrices.<n>We prove that the dual gap of the greedy policy is bounded by the square of the estimation error.
  arXiv  Detail & Related papers  (2026-02-26T15:27:53Z)
• Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs [55.77845440440496]
  Push-based decentralized communication enables optimization over communication networks, where information exchange may be asymmetric.<n>We develop a unified uniform-stability framework for the Gradient Push (SGP) algorithm.<n>A key technical ingredient is an imbalance-aware generalization bound through two quantities.
  arXiv  Detail & Related papers  (2026-02-24T05:32:03Z)
• MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning [16.012761588513026]
  Reinforcement Learning with Verifiable Rewards (RLVR) algorithms rely on rigid, uniform, and symmetric trust region mechanisms.<n>We propose Mass-Adaptive Soft Policy Optimization (MASPO), a unified framework designed to harmonize these three dimensions.<n> MASPO integrates a differentiable soft Gaussian gating to maximize gradient utility, a mass-adaptive limiter to balance exploration across the probability spectrum, and an asymmetric risk controller to align update magnitudes with signal confidence.
  arXiv  Detail & Related papers  (2026-02-19T17:05:20Z)
• Row-stochastic matrices can provably outperform doubly stochastic matrices in decentralized learning [10.686669655748702]
  Decentralized learning often involves a weighted global loss with heterogeneous node weights $$.<n>We develop a weighted Hilbert-space framework $L2(mathbbRd)$ and obtain convergence rates that are strictly tighter than those from Euclidean analysis.<n>We then derive sufficient conditions under which the row-stochastic design converges faster even with a smaller spectral gap.
  arXiv  Detail & Related papers  (2025-11-24T02:58:38Z)
• Graph-based Clustering Revisited: A Relaxation of Kernel $k$-Means Perspective [73.18641268511318]
  We propose a graph-based clustering algorithm that only relaxes the orthonormal constraint to derive clustering results.<n>To ensure a doubly constraint into a gradient, we transform the non-negative constraint into a class probability parameter.
  arXiv  Detail & Related papers  (2025-09-23T09:14:39Z)
• NDCG-Consistent Softmax Approximation with Accelerated Convergence [67.10365329542365]
  We propose novel loss formulations that align directly with ranking metrics.<n>We integrate the proposed RG losses with the highly efficient Alternating Least Squares (ALS) optimization method.<n> Empirical evaluations on real-world datasets demonstrate that our approach achieves comparable or superior ranking performance.
  arXiv  Detail & Related papers  (2025-06-11T06:59:17Z)
• An Accelerated Alternating Partial Bregman Algorithm for ReLU-based Matrix Decomposition [0.0]
  In this paper, we aim to investigate the sparse low-rank characteristics rectified on non-negative matrices.<n>We propose a novel regularization term incorporating useful structures in clustering and compression tasks.<n>We derive corresponding closed-form solutions while maintaining the $L$-smooth property always holds for any $Lge 1$.
  arXiv  Detail & Related papers  (2025-03-04T08:20:34Z)
• Distributional Robustness and Regularization in Reinforcement Learning [62.23012916708608]
  We introduce a new regularizer for empirical value functions and show that it lower bounds the Wasserstein distributionally robust value function. It suggests using regularization as a practical tool for dealing with $textitexternal uncertainty$ in reinforcement learning.
  arXiv  Detail & Related papers  (2020-03-05T19:56:23Z)

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.