paper-with-me

홈 › Papers

Fast, Small and Exact: Infinite-order Language Modelling with Compressed Suffix Trees

2016-08-16 · TACL 2016 1 · Ehsan Shareghi, Matthias Petri, Gholamreza Haffari, Trevor Cohn

Efficient methods for storing and querying are critical for scaling high-order n-gram language models to large corpora. We propose a language model based on compressed suffix trees, a representation that is highly compact and can be easily held in memory, while supporting queries needed in computing language model probabilities on-the-fly. We present several optimisations which improve query runtimes up to 2500x, despite only incurring a modest increase in construction time and memory usage. For large corpora and high Markov orders, our method is highly competitive with the state-of-the-art KenLM package. It imposes much lower memory requirements, often by orders of magnitude, and has runtimes that are either similar (for training) or comparable (for querying).

📄 PDF Abstract BibTeX arXiv:1608.04465

Code (1)

eehsan/cstlm 공식 구현

Tasks

Language ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

On Noisy Negative Curvature Descent: Competing with Gradient Descent for Faster Non-convex Optimization

2017-09-25 · Mingrui Liu, Tianbao Yang

The Hessian-vector product has been utilized to find a second-order stationary solution with strong complexity guarantee (e.g., almost linear time complexity in the problem's dimensionality). In this paper, we propose to…

The Method of Infinite Descent

2025-10-07 · Reza T. Batley, Sourav Saha arxiv

Training - the optimisation of complex models - is traditionally performed through small, local, iterative updates [D. E. Rumelhart, G. E. Hinton, R. J. Williams, Nature 323, 533-536 (1986)]. Approximating solutions thro…

Application of MUSIC-type imaging for anomaly detection without background information

2023-07-03 · Won-Kwang Park

It has been demonstrated that the MUltiple SIgnal Classification (MUSIC) algorithm is fast, stable, and effective for localizing small anomalies in microwave imaging. For the successful application of MUSIC, exact values…

Anomaly Detection

SR-TTT: Surprisal-Aware Residual Test-Time Training

2026-02-26 · Swamynathan V P arxiv

Test-Time Training (TTT) language models achieve theoretically infinite context windows with an O(1) memory footprint by replacing the standard exact-attention KV-cache with hidden state ``fast weights'' W_fast updated v…

Self-Supervised Learning

Probabilistic Inference with Generating Functions for Poisson Latent Variable Models

2016-12-01 · NeurIPS 2016 12 · Kevin Winner, Daniel R. Sheldon

Graphical models with latent count variables arise in a number of fields. Standard exact inference techniques such as variable elimination and belief propagation do not apply to these models because the latent variables …