Late-Stage Generalization Collapse in Grokking: Detecting anti-grokking with Weightwatcher

• URL: http://arxiv.org/abs/2602.02859v1
• Date: Mon, 02 Feb 2026 22:09:14 GMT
• Title: Late-Stage Generalization Collapse in Grokking: Detecting anti-grokking with Weightwatcher
• Abstract summary: emphMemorization in neural networks lacks a precise operational definition and is often inferred from the grokking regime.<n>We identify a previously unreported third phase of grokking in this training regime: emphanti-grokking, a late-stage collapse of generalization.
• Score: 1.6615337656760856
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: \emph{Memorization} in neural networks lacks a precise operational definition and is often inferred from the grokking regime, where training accuracy saturates while test accuracy remains very low. We identify a previously unreported third phase of grokking in this training regime: \emph{anti-grokking}, a late-stage collapse of generalization. We revisit two canonical grokking setups: a 3-layer MLP trained on a subset of MNIST and a transformer trained on modular addition, but extended training far beyond standard. In both cases, after models transition from pre-grokking to successful generalization, test accuracy collapses back to chance while training accuracy remains perfect, indicating a distinct post-generalization failure mode. To diagnose anti-grokking, we use the open-source \texttt{WeightWatcher} tool based on HTSR/SETOL theory. The primary signal is the emergence of \emph{Correlation Traps}: anomalously large eigenvalues beyond the Marchenko--Pastur bulk in the empirical spectral density of shuffled weight matrices, which are predicted to impair generalization. As a secondary signal, anti-grokking corresponds to the average HTSR layer quality metric $α$ deviating from $2.0$. Neither metric requires access to the test or training data. We compare these signals to alternative grokking diagnostic, including $\ell_2$ norms, Activation Sparsity, Absolute Weight Entropy, and Local Circuit Complexity. These track pre-grokking and grokking but fail to identify anti-grokking. Finally, we show that Correlation Traps can induce catastrophic forgetting and/or prototype memorization, and observe similar pathologies in large-scale LLMs, like OSS GPT 20/120B.

Related papers

• Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems [47.027290803102666]
  Normal behaviour is the union of many imbalanced, curved, thin-fringed operating regimes.<n>We model the normal law with a jointly learned latent representation plus explicit Gaussian-mixture mode clustering.<n>We evaluate under a deliberately fair protocol: raw point-wise metrics with no point adjustment, a trivial-detector difficulty split, prevalence-matched F1, and train-normal-only calibration.
  arXiv  Detail & Related papers  (2026-07-07T10:10:00Z)
• Token-Level Generalization in LoRA Adapter Backdoors: Attack Characterization and Behavioral Detection [0.0]
  We show that LoRA can be reliably backdoored through training data poisoning.<n>The attack scales monotonically with rank, and the chosen trigger-anchor token is both trigger-dependent and base-model-dependent.<n> Behavioral detection is the operationally portable result for supply chain scanning.
  arXiv  Detail & Related papers  (2026-05-28T16:32:25Z)
• Detecting overfitting in Neural Networks during long-horizon grokking using Random Matrix Theory [1.6615337656760856]
  We present a novel Random Matrix Theory method that detects the onset of overfitting in deep learning models without access to train or test data.<n>During the onset of overfitting, which we call the "anti-grokking" phase in long-horizon grokking, Correlation Traps form and grow in number and scale.<n>Our findings show that anti-grokking is an additional grokking phase with high train accuracy and decreasing test accuracy, structurally distinct from pre-grokking through its Correlation Traps.
  arXiv  Detail & Related papers  (2026-05-12T16:57:48Z)
• Distributional Spectral Diagnostics for Localizing Grokking Transitions [4.480801209867367]
  In grokking, a model first fits the training data while test accuracy remains low, and only later begins to generalize.<n>We ask whether this transition can be localized from observed training trajectories before the test accuracy rises, and formulate grokking transition localization as a diagnostic problem with an explicit threshold/FPR/lead-time trade-off.
  arXiv  Detail & Related papers  (2026-05-07T03:08:58Z)
• 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)
• Sparse Semantic Dimension as a Generalization Certificate for LLMs [53.681678236115836]
  We introduce the Sparse Semantic Dimension (SSD), a complexity measure derived from the active feature vocabulary of a Sparse Autoencoder (SAE) trained on the model's layers.<n>We validate this framework on GPT-2 Small and Gemma-2B, demonstrating that our bound provides non-vacuous certificates at realistic sample sizes.
  arXiv  Detail & Related papers  (2026-02-11T21:45:18Z)
• Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol [69.11739400975445]
  We introduce the first theoretical framework for analyzing error accumulation in Model Context Protocol (MCP) agents.<n>We show that cumulative distortion exhibits linear growth and high-probability deviations bounded by $O(sqrtT)$.<n>Key findings include: semantic weighting reduces distortion by 80%, and periodic re-grounding approximately every 9 steps suffices for error control.
  arXiv  Detail & Related papers  (2026-02-10T21:08:53Z)
• Thinking Traps in Long Chain-of-Thought: A Measurable Study and Trap-Aware Adaptive Restart [27.904791075662896]
  We introduce TAAR (Trap-Aware Adaptive Restart), a test-time control framework that trains a diagnostic policy to predict two signals from partial trajectories.<n>At inference time, TAAR truncates the trajectory before the predicted trap segment and adaptively restarts decoding.<n>Experiments show that TAAR improves reasoning performance without fine-tuning base model parameters.
  arXiv  Detail & Related papers  (2026-01-17T07:26:02Z)
• Revisiting Weighted Strategy for Non-stationary Parametric Bandits and MDPs [56.246783503873225]
  This paper revisits the weighted strategy for non-stationary parametric bandits.<n>We propose a simpler weight-based algorithm that is as efficient as window/restart-based algorithms.<n>Our framework can be used to improve regret bounds of other parametric bandits.
  arXiv  Detail & Related papers  (2026-01-03T04:50:21Z)
• Phase-space entropy at acquisition reflects downstream learnability [54.4100065023873]
  We propose an acquisition-level scalar $S_mathcal B$ based on instrument-resolved phase space.<n>We show theoretically that (S_mathcal B) correctly identifies the phase-space coherence of periodic sampling.<n>$|S_mathcal B|$ consistently ranks sampling geometries and predicts downstream reconstruction/recognition difficulty emphwithout training.
  arXiv  Detail & Related papers  (2025-12-22T10:03:51Z)
• Adapt in the Wild: Test-Time Entropy Minimization with Sharpness and Feature Regularization [85.50560211492898]
  Test-time adaptation (TTA) may fail to improve or even harm the model performance when test data have mixed distribution shifts.<n>This is often a key obstacle preventing existing TTA methods from being deployed in the real world.<n>We propose a sharpness-aware and reliable entropy minimization method, called SAR, for stabilizing TTA from two aspects.
  arXiv  Detail & Related papers  (2025-09-05T10:03:00Z)
• Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learning [12.690752260618522]
  Confident Meta-learning (CoMet) is a novel training strategy that enables deep anomaly detection models to learn from uncurated datasets.<n>Our approach integrates Soft Confident Learning, which assigns lower weights to low-confidence samples, and Meta-Learning, which stabilizes training by regularizing updates.
  arXiv  Detail & Related papers  (2025-08-04T11:03:12Z)
• Grokking and Generalization Collapse: Insights from \texttt{HTSR} theory [4.371245698184159]
  We study the well-known grokking phenomena in neural networks (NNs) using a 3-layer trained on 1 k-sample subset of MNIST.<n>We discover a novel third phase -- emphanti-grokking -- that occurs very late in training and resembles but is distinct from the familiar emphpre-grokking phases.
  arXiv  Detail & Related papers  (2025-06-04T20:34:37Z)
• Two failure modes of deep transformers and how to avoid them: a unified theory of signal propagation at initialisation [8.973965016201822]
  Finding the right initialisation for neural networks is crucial to ensure smooth training and good performance.<n>In transformers, the wrong initialisation can lead to one of two failure modes of self-attention layers: rank collapse, where all tokens collapse into similar representations, and entropy collapse, where highly concentrated attention scores lead to instability.<n>Here, we provide an analytical theory of signal propagation through deep transformers with self-attention, layer normalisation, skip connections and gradients.
  arXiv  Detail & Related papers  (2025-05-30T08:18:23Z)
• Beyond Progress Measures: Theoretical Insights into the Mechanism of Grokking [50.465604300990904]
  Grokking refers to the abrupt improvement in test accuracy after extended overfitting.<n>In this work, we investigate the grokking mechanism underlying the Transformer in the task of prime number operations.
  arXiv  Detail & Related papers  (2025-04-04T04:42:38Z)
• Regularized Linear Regression for Binary Classification [20.710343135282116]
  Regularized linear regression is a promising approach for binary classification problems in which the training set has noisy labels. We show that for large enough regularization strength, the optimal weights concentrate around two values of opposite sign. We observe that in many cases the corresponding "compression" of each weight to a single bit leads to very little loss in performance.
  arXiv  Detail & Related papers  (2023-11-03T23:18:21Z)
• Industrial Anomaly Detection and Localization Using Weakly-Supervised Residual Transformers [44.344548601242444]
  We introduce a novel framework, Weakly-supervised RESidual Transformer (WeakREST), to achieve high anomaly detection accuracy.<n>We reformulate the pixel-wise anomaly localization task into a block-wise classification problem.<n>We develop a novel ResMixMatch algorithm, capable of handling the interplay between weak labels and residual-based representations.
  arXiv  Detail & Related papers  (2023-06-06T08:19:30Z)
• Simple and Effective Prevention of Mode Collapse in Deep One-Class Classification [93.2334223970488]
  We propose two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small.
  arXiv  Detail & Related papers  (2020-01-24T03:44:47Z)

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.