Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers

• URL: http://arxiv.org/abs/2607.04819v2
• Date: Mon, 13 Jul 2026 09:12:14 GMT
• Title: Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers
• Abstract summary: We study whether Structured Layerism (SNLP) can make this inter-layer composition more FHE-friendly.<n>SNLP reduces the layerwise nonlinear depth from L stages to a small number of solver plus linear structured corrections.<n>Across all tested models, SNLP has lower amplification than sequential iterations.
• Score: 25.98504186362678
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Fully homomorphic encryption (FHE) enables computation on encrypted data, but practical encrypted Transformer inference is bottlenecked by the sequential composition of many nonlinear blocks. We study whether Structured Newton Layer Parallelism (SNLP) can make this inter-layer composition more FHE-friendly: each Transformer block still requires polynomial approximations for operations such as softmax and RMSNorm, but SNLP reduces the layerwise sequential nonlinear depth from L stages to a small number of solver iterations plus linear structured corrections. Using a simulation framework based on Chebyshev polynomial approximations, we measure error accumulation under sequential versus SNLP inference across 8 models and 4 architecture families. On a 0.5B IDN-trained model, SNLP reduces symbolic bootstraps from 53 to 20 (2.65x) with only +1.2% perplexity degradation, while lowering error amplification (1.36x vs. 1.42x). Across all tested models, SNLP has lower amplification than sequential inference. Ablations show that softmax approximation dominates the error budget and CKKS arithmetic noise is negligible in our setting, suggesting that SNLP is complementary to block-level FHE-friendly operator design rather than a replacement for it.

Related papers

• Per-Matrix Optimality Is Not Enough: Three-Level Optimization for Low-Rank LLM Compression [8.890261387138215]
  We introduce a three-level chain that widens optimization scope from individual matrices to Transformer blocks to the full model.<n>On LLaMA-7B at 60% compression, the chain reduces WikiText-2 perplexity from 42.1 to 19.1 to 11.4.<n> skipping the block-level stage becomes competitive, revealing an offline compute--data trade-off.
  arXiv  Detail & Related papers  (2026-09-14T16:36:46Z)
• Structured Transforms for Low-Overhead Quantization of Language Models [43.017319901277354]
  We revisit Kashin-decomposition-based weight quantization for large language models and propose an improved algorithm.<n>The proposed greedy algorithm with alternating updates guarantees the four-peak distribution required for stable 2-bit clustering of each factor.<n>The resulting JAX pipeline is competitive with OPTQ, QuIP, QuIP-RG and a fine-tuning- and vector-quantization-free variant of QuIP# at 4-bit per channel on OPT, Llama-2 and Pythia.
  arXiv  Detail & Related papers  (2026-09-10T15:15:50Z)
• Cross-Layer Error Compensation and Finite-Sample Feature-Statistics Matching for Extreme Low-Bit Quantization of Large Language Models [0.0]
  We formulate quantization as a joint optimization over the discrete codes and scales of all layers.<n>We show that error compensation alone reaches a perplexity ratio of 9.56 +/- 0.15 over the FP16 teacher.<n>On Qwen2.5-1.5B with 1.125-bit group-binary weights, error compensation alone reaches a perplexity ratio of 9.56 +/- 0.15 over the FP16 teacher.
  arXiv  Detail & Related papers  (2026-07-16T06:47:52Z)
• Hidden Decoding at Scale: Latent Computation Scaling for Large Language Models [75.80275843320511]
  We study whether an existing backbone can keep improving by allocating more computation to each token while leaving the Transformer backbone fixed.<n>We propose Hidden Decoding, a sequence-length scaling method applied during continued pretraining.
  arXiv  Detail & Related papers  (2026-07-09T07:37:59Z)
• Closed-Form Spectral Regularization for Multi-Task Model Merging [96.82449201305234]
  Model merging combines several independently fine-tuned experts into a single multi-task model without any training data.<n>State-of-the-art merging methods formulate merging as a layer-wise interference problem.<n>We propose SWUDI, a closed-form method that combines a soft exponential filter, which matches the gradient-flow trajectory of iterative descent.
  arXiv  Detail & Related papers  (2026-06-05T14:00:47Z)
• SNLP: Layer-Parallel Inference via Structured Newton Corrections [22.126763421836966]
  We study whether layerwise dependency can be relaxed by treating the hidden-state trace across layers as the solution of a nonlinear residual equation.<n>We introduce Structured Newton Layer Parallelism (SNLP), a training and inference framework that replaces exact layer Jacobians with cheap architecture-induced surrogate dynamics.
  arXiv  Detail & Related papers  (2026-05-18T04:28:16Z)
• Normalized Architectures are Natively 4-Bit [49.13186675123547]
  nGPT, an architecture that constrains weights and hidden representations to the unit hypersphere, is inherently more robust to low-precision arithmetic.<n>We validate this approach on both a 1.2B dense model and hybrid (Mamba-Transformer) MoE models of up to 3B/30B parameters.
  arXiv  Detail & Related papers  (2026-05-07T11:54:07Z)
• MLPMoE: Zero-Shot Architectural Metamorphosis of Dense LLM MLPs into Static Mixture-of-Experts [0.0]
  Large Language Models (LLMs) are predominantly deployed as dense transformers, where every parameter in every feed-forward block is activated for every token.<n>Recent upcycling methods such as MoEfication, CMoE, ToMoE, and MoORE reveal that much of the useful computation lives in sparse, semi-modular substructures inside dense feed-forward networks.<n>This paper introducesMoE (MLP-Experts), a training-free transformation that restructures the dense in transformer blocks into a static, high-cardinality mixture of experts.
  arXiv  Detail & Related papers  (2025-11-26T06:14:26Z)
• Don't Be Greedy, Just Relax! Pruning LLMs via Frank-Wolfe [61.68406997155879]
  State-of-the-art Large Language Model (LLM) pruning methods operate layer-wise, minimizing the per-layer pruning error on a small dataset to avoid full retraining.<n>Existing methods hence rely on greedy convexs that ignore the weight interactions in the pruning objective.<n>Our method drastically reduces the per-layer pruning error, outperforms strong baselines on state-of-the-art GPT architectures, and remains memory-efficient.
  arXiv  Detail & Related papers  (2025-10-15T16:13:44Z)
• PAPER: Privacy-Preserving ResNet Models using Low-Degree Polynomial Approximations and Structural Optimizations on Leveled FHE [5.819818547073678]
  Recent work has made non-interactive privacy-preserving inference more practical by running deep Convolution Neural Network (CNN) with Fully Homomorphic Encryption (FHE)<n>They also depend on high-degree approximations of non-linear activations, which increase multiplicative depth and reduce accuracy by 2-5% compared to plaintext ReLU models.<n>In this work, we focus on ResNets, a widely adopted benchmark architecture in privacy-preserving inference, and close the accuracy gap between their FHE non-interactive models and counterparts.
  arXiv  Detail & Related papers  (2025-09-26T19:10:23Z)
• Efficient Large Language Model Inference with Neural Block Linearization [51.619870789584525]
  We introduce Neural Block Linearization (NBL), a novel framework for accelerating transformer model inference.<n>NBL replaces self-attention layers with linear approximations derived from Linear Minimum Mean Squared Error estimators.<n>In experiments, NBL achieves notable computational speed-ups while preserving competitive accuracy.
  arXiv  Detail & Related papers  (2025-05-27T12:01:43Z)
• KHRONOS: a Kernel-Based Neural Architecture for Rapid, Resource-Efficient Scientific Computation [0.9355993154058798]
  We introduce KHRONOS, an AI framework for model based, model free and model inversion tasks.<n>KHRONOS constructs continuously differentiable target fields with a hierarchical composition of per-dimension kernel expansions.<n>For inverse problems, KHRONOS facilitates rapid, iterative level set recovery in only a few forward evaluations, with sub-microsecond per sample latency.
  arXiv  Detail & Related papers  (2025-05-19T16:29:07Z)
• Factorizers for Distributed Sparse Block Codes [45.29870215671697]
  We propose a fast and highly accurate method for factorizing distributed block codes (SBCs) Our iterative factorizer introduces a threshold-based nonlinear activation, conditional random sampling, and an $ell_infty$-based similarity metric. We demonstrate the feasibility of our method on four deep CNN architectures over CIFAR-100, ImageNet-1K, and RAVEN datasets.
  arXiv  Detail & Related papers  (2023-03-24T12:31:48Z)
• Improved techniques for deterministic l2 robustness [63.34032156196848]
  Training convolutional neural networks (CNNs) with a strict 1-Lipschitz constraint under the $l_2$ norm is useful for adversarial robustness, interpretable gradients and stable training. We introduce a procedure to certify robustness of 1-Lipschitz CNNs by replacing the last linear layer with a 1-hidden layer. We significantly advance the state-of-the-art for standard and provable robust accuracies on CIFAR-10 and CIFAR-100.
  arXiv  Detail & Related papers  (2022-11-15T19:10:12Z)

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.