Beyond N-gram: Data-Aware X-GRAM Extraction for Efficient Embedding Parameter Scaling

• URL: http://arxiv.org/abs/2604.21724v2
• Date: Fri, 24 Apr 2026 17:56:56 GMT
• Title: Beyond N-gram: Data-Aware X-GRAM Extraction for Efficient Embedding Parameter Scaling
• Abstract summary: X-GRAM is a frequency-aware dynamic token-injection framework.<n>It employs hybrid hashing and alias mixing to compress the tail while preserving head capacity.<n>It refines retrieved vectors via normalized SwiGLU ShortConv to extract diverse local n-gram features.
• Score: 24.48256429931821
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large token-indexed lookup tables provide a compute-decoupled scaling path, but their practical gains are often limited by poor parameter efficiency and rapid memory growth. We attribute these limitations to Zipfian under-training of the long tail, heterogeneous demand across layers, and "slot collapse" that produces redundant embeddings. To address this, we propose X-GRAM, a frequency-aware dynamic token-injection framework. X-GRAM employs hybrid hashing and alias mixing to compress the tail while preserving head capacity, and refines retrieved vectors via normalized SwiGLU ShortConv to extract diverse local n-gram features. These signals are integrated into attention value streams and inter-layer residuals using depth-aware gating, effectively aligning static memory with dynamic context. This design introduces a memory-centric scaling axis that decouples model capacity from FLOPs. Extensive evaluations at the 0.73B and 1.15B scales show that X-GRAM improves average accuracy by as much as 4.4 points over the vanilla backbone and 3.2 points over strong retrieval baselines, while using substantially smaller tables in the 50% configuration. Overall, by decoupling capacity from compute through efficient memory management, X-GRAM offers a scalable and practical paradigm for future memory-augmented architectures. Code aviliable in https://github.com/Longyichen/X-gram.

Related papers

• FluxBin: Flexible LUT-based Ultra-low-bit LLM Inference by Algorithm-Kernel Synergy [32.21687694159134]
  FluxBin is an algorithm that synerizes binary quantization with a highly optimized kernel.<n>FluxBin achieves up to $5.92times speedup and $10.19times energy savings.<n>This effectively enables the deployment of 70B-scale models on one single A100 bases with a $4times memory reduction.
  arXiv  Detail & Related papers  (2026-08-16T08:01:06Z)
• FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention [77.12062766962815]
  Lookahead Sparse Attention (LSA) is powered by a Neural Memory Indexer built upon the DeepSeek-V4 architecture.<n>We instantiate this architecture via a backbone-free decoupled training strategy.<n>We show that FM-DS-V4 compresses the average physical KV cache footprint down to merely 13.5% of the full-context baseline.
  arXiv  Detail & Related papers  (2026-06-08T06:25:54Z)
• LimiX-2M: Mitigating Low-Rank Collapse and Attention Bottlenecks in Tabular Foundation Models [56.999481798138625]
  LimiX-2M is a 2M- parameter model that outperforms larger TabPFN-v2 and TabICL baselines on widely used benchmarks.<n>We present a unified tokenize-and-route framework for strong Tabular foundation models (TFMs)<n>Results highlight value-aware tokenization and readout-aligned routing as key levers for improving the accuracy--efficiency trade-off in TFMs.
  arXiv  Detail & Related papers  (2026-06-03T06:07:33Z)
• HASTE: Hardware-Aware Dynamic Sparse Training for Large Output Spaces [11.730547294888707]
  Extreme multi-label classification (XMC) involves learning models over large output spaces with millions of labels.<n>Group-shared fixed fan-in sparsity is a semi-structured output-layer design in which semantically related labels share a sparse input pattern.<n>We show that group-shared fixed fan-in arithmetic reduces into practical wall-clock gains.
  arXiv  Detail & Related papers  (2026-05-31T09:25:47Z)
• Memory Grafting: Scaling Language Model Pre-training via Offline Conditional Memory [65.39827296429527]
  Scaling conditional memory offers a promising way to increase language-model capacity.<n>Existing methods such as Engram learn large memory tables from scratch during pre-training.<n>We propose Memory Grafting, a conditional memory scaling method that utilizes frozen hidden states from a grafting model as conditional n-gram memory.
  arXiv  Detail & Related papers  (2026-05-20T09:35:04Z)
• OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond [50.440302567029654]
  Multi-modal intelligence has made the Key-Value cache a dominant memory bottleneck for efficient deployment.<n>In this work, we revisit the inherent limitations of the per-channel quantization paradigm.<n>We propose OScaR, an accurate and lightweight KV cache compression framework for X-LLMs.
  arXiv  Detail & Related papers  (2026-05-19T10:53:03Z)
• CATS: Cascaded Adaptive Tree Speculation for Memory-Limited LLM Inference Acceleration [2.6196419303092466]
  Auto-regressive decoding in Large Language Models (LLMs) is inherently memory-bound.<n>We propose CATS, a self-speculative decoding framework that conducts cascaded verification and correction on memory-limited devices.<n> CATS can achieve a wall-clock speedup of up to 5.08x with no degradation in generation quality, outperforming the SOTA method by up to 1.45x under edge memory constraints.
  arXiv  Detail & Related papers  (2026-05-11T19:50:08Z)
• XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compression [20.18561757219652]
  XStreamVGGT is a tuning-free approach that seamlessly integrates pruning and quantization to systematically compress the KV cache.<n>XStreamVGGT achieves mostly negligible performance degradation while substantially reducing memory usage by 4.42$times$.
  arXiv  Detail & Related papers  (2026-02-25T11:02:02Z)
• Hierarchical Adaptive Eviction for KV Cache Management in Multimodal Language Models [8.944739362562494]
  Existing KV cache eviction strategies fail to address the heterogeneous attention distributions between visual and text tokens.<n>We propose Hierarchical Adaptive Eviction (HAE), a KV cache eviction framework that optimize text-visual token interaction in MLLMs.<n>HAE minimizes KV cache usage across layers, reduces computational overhead via index broadcasting, and theoretically ensures superior information integrity and lower error bounds.
  arXiv  Detail & Related papers  (2026-02-02T15:01:44Z)
• Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models [42.816060150754645]
  We introduce conditional memory as a complementary sparsity axis, instantiated via Engram, a module that modernizes classic $N$-gram embedding for O(1) lookup.<n>We scale Engram to 27B parameters, achieving superior performance over a strictly iso- parameter and iso-FLOPs MoE baseline.<n>We envision conditional memory as an indispensable modeling primitive for next-generation sparse models.
  arXiv  Detail & Related papers  (2026-01-12T09:54:49Z)
• XQuant: Achieving Ultra-Low Bit KV Cache Quantization with Cross-Layer Compression [54.28208936996186]
  Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks.<n> Quantization has emerged as a promising solution to reduce memory consumption while preserving historical information.<n>We propose XQuant, a training-free and plug-and-play framework that achieves ultra-low equivalent bit-width KV cache quantization.
  arXiv  Detail & Related papers  (2025-10-13T10:17:21Z)
• Efficient In-Memory Acceleration of Sparse Block Diagonal LLMs [1.5099440840283813]
  Structured sparsity enables deploying large language models (LLMs) on resource-constrained systems.<n>We present a framework with novel mapping and scheduling strategies to accelerate sparse LLM inference on CIM accelerators.<n>Our approach improves CIM utilization by over 50%, achieving more than 4x reduction in both memory footprint and the number of required floating-point operations.
  arXiv  Detail & Related papers  (2025-10-13T09:25:48Z)
• Re-Densification Meets Cross-Scale Propagation: Real-Time Neural Compression of LiDAR Point Clouds [83.39320394656855]
  LiDAR point clouds are fundamental to various applications, yet high-precision scans incur substantial storage and transmission overhead.<n>Existing methods typically convert unordered points into hierarchical octree or voxel structures for dense-to-sparse predictive coding.<n>Our framework comprises two lightweight modules. First, the Geometry Re-Densification Module re-densifies encoded sparse geometry, extracts features at denser scale, and then re-sparsifies the features for predictive coding.
  arXiv  Detail & Related papers  (2025-08-28T06:36:10Z)
• A Universal Framework for Compressing Embeddings in CTR Prediction [68.27582084015044]
  We introduce a Model-agnostic Embedding Compression (MEC) framework that compresses embedding tables by quantizing pre-trained embeddings.<n>Our approach consists of two stages: first, we apply popularity-weighted regularization to balance code distribution between high- and low-frequency features.<n> Experiments on three datasets reveal that our method reduces memory usage by over 50x while maintaining or improving recommendation performance.
  arXiv  Detail & Related papers  (2025-02-21T10:12:34Z)
• ThinK: Thinner Key Cache by Query-Driven Pruning [63.13363917871414]
  Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications.<n>This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference.<n>We propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels.
  arXiv  Detail & Related papers  (2024-07-30T17:59:08Z)
• NumS: Scalable Array Programming for the Cloud [82.827921577004]
  We present NumS, an array programming library which optimize NumPy-like expressions on task-based distributed systems. This is achieved through a novel scheduler called Load Simulated Hierarchical Scheduling (LSHS) We show that LSHS enhances performance on Ray by decreasing network load by a factor of 2x, requiring 4x less memory, and reducing execution time by 10x on the logistic regression problem.
  arXiv  Detail & Related papers  (2022-06-28T20:13:40Z)

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.