This paper introduces a flexible Transformer-based model for detecting anomalies in system logs. By embedding log templates with a pre-trained BERT model and incorporating positional and temporal encoding, it captures both semantic and sequential context within log sequences. The approach supports variable sequence lengths and configurable input features, enabling extensive experimentation across datasets. The model performs supervised binary classification to distinguish normal from anomalous patterns, using a [CLS]-like token for sequence-level representation. Overall, it pushes the boundaries of log-based anomaly detection by integrating modern NLP and deep learning techniques into system monitoring.This paper introduces a flexible Transformer-based model for detecting anomalies in system logs. By embedding log templates with a pre-trained BERT model and incorporating positional and temporal encoding, it captures both semantic and sequential context within log sequences. The approach supports variable sequence lengths and configurable input features, enabling extensive experimentation across datasets. The model performs supervised binary classification to distinguish normal from anomalous patterns, using a [CLS]-like token for sequence-level representation. Overall, it pushes the boundaries of log-based anomaly detection by integrating modern NLP and deep learning techniques into system monitoring.

Transformer-Based Anomaly Detection Using Log Sequence Embeddings

2025/11/04 01:52

Abstract

1 Introduction

2 Background and Related Work

2.1 Different Formulations of the Log-based Anomaly Detection Task

2.2 Supervised v.s. Unsupervised

2.3 Information within Log Data

2.4 Fix-Window Grouping

2.5 Related Works

3 A Configurable Transformer-based Anomaly Detection Approach

3.1 Problem Formulation

3.2 Log Parsing and Log Embedding

3.3 Positional & Temporal Encoding

3.4 Model Structure

3.5 Supervised Binary Classification

4 Experimental Setup

4.1 Datasets

4.2 Evaluation Metrics

4.3 Generating Log Sequences of Varying Lengths

4.4 Implementation Details and Experimental Environment

5 Experimental Results

5.1 RQ1: How does our proposed anomaly detection model perform compared to the baselines?

5.2 RQ2: How much does the sequential and temporal information within log sequences affect anomaly detection?

5.3 RQ3: How much do the different types of information individually contribute to anomaly detection?

6 Discussion

7 Threats to validity

8 Conclusions and References

\

3 A Configurable Transformer-based Anomaly Detection Approach

In this study, we introduce a novel transformer-based method for anomaly detection. The model takes log sequences as inputs to detect anomalies. The model employs a pretrained BERT model to embed log templates, enabling the representation of semantic information within log messages. These embeddings, combined with positional or temporal encoding, are subsequently inputted into the transformer model. The combined information is utilized in the subsequent generation of log sequence-level representations, facilitating the anomaly detection process. We design our model to be flexible: The input features are configurable so that we can use or conduct experiments with different feature combinations of the log data. Additionally, the model is designed and trained to handle input log sequences of varying lengths. In this section, we introduce our problem formulation and the detailed design of our method.

\ 3.1 Problem Formulation

We follow the previous works [1] to formulate the task as a binary classification task, in which we train our proposed model to classify log sequences into anomalies and normal ones in a supervised way. For the samples used in the training and evaluation of the model, we utilize a flexible grouping approach to generate log sequences of varying lengths. The details are introduced in Section 4

\ 3.2 Log Parsing and Log Embedding

In our work, we transform log events into numerical vectors by encoding log templates with a pre-trained language model. To obtain the log templates, we adopt the Drain parser [24], which is widely used and has good parsing performance on most of the public datasets [4]. We use a pre-trained sentence-bert model [25] (i.e., all-MiniLML6-v2 [26]) to embed the log templates generated by the log parsing process. The pre-trained model is trained with a contrastive learning objective and achieves state-ofthe-art performance on various NLP tasks. We utilize this pre-trained model to create a representation that captures semantic information of log messages and illustrates the similarity between log templates for the downstream anomaly detection model. The output dimension of the model is 384.

\ 3.3 Positional & Temporal Encoding

The original transformer model [27] adopts a positional encoding to enable the model to make use of the order of the input sequence. As the model contains no recurrence and no convolution, the models will be agnostic to the log sequence without the positional encoding. While some studies suggest that transformer models without explicit positional encoding remain competitive with standard models when dealing with sequential data [28, 29], it is important to note that any permutation of the input sequence will produce the same internal state of the model. As sequential information or temporal information may be important indicators for anomalies within log sequences, previous works that are based on transformer models utilize the standard positional encoding to inject the order of log events or templates in the sequence [11, 12, 21], aiming to detect anomalies associated with the wrong execution order. However, we noticed that in a common-used replication implementation of a transformer-based method [5], the positional encoding was, in fact, omitted. To the best of our knowledge, no existing work has encoded the temporal information based on the timestamps of logs for their anomaly detection method. The effectiveness of utilizing sequential or temporal information in the anomaly detection task is unclear.

\ In our proposed method, we attempt to incorporate sequential and temporal encoding into the transformer model and explore the importance of sequential and temporal information for anomaly detection. Specifically, our proposed method has different variants utilizing the following sequential or temporal encoding techniques. The encoding is then added to the log representation, which serves as the input to the transformer structure.

\

3.3.1 Relative Time Elapse Encoding (RTEE)

We propose this temporal encoding method, RTEE, which simply substitutes the position index in positional encoding with the timing of each log event. We first calculate the time elapse according to the timestamps of log events in the log sequence. Instead of using the log event sequence index as the position to sinusoidal and cosinusoidal equations, we use the relative time elapse to the first log event in the log sequence to substitute the position index. Table 1 shows an example of time intervals in a log sequence. In the example, we have a log sequence containing 7 events with a time span of 7 seconds. The elapsed time from the first event to each event in the sequence is utilized to calculate the time encoding for the corresponding events. Similar to positional encoding, the encoding is calculated with the above-mentioned equations 1, and the encoding will not update during the training process.

\

3.4 Model Structure

The transformer is a neural network architecture that relies on the self-attention mechanism to capture the relationship between input elements in a sequence. The transformer-based models and frameworks have been used in the anomaly detection task by many previous works [6, 11, 12, 21]. Inspired by the previous works, we use a transformer encoder-based model for anomaly detection. We design our approach to accept log sequences of varying lengths and generate sequence-level representations. To achieve this, we have employed some specific tokens in the input log sequence for the model to generate sequence representation and identify the padded tokens and the end of the log sequence, drawing inspiration from the design of the BERT model [31]. In the input log sequence, we used the following tokens: is placed at the start of each sequence to allow the model to generate aggregated information for the entire sequence, is added at the end of the sequence to signify its completion, is used to mark the masked tokens under the self-supervised training paradigm, and is used for padded tokens. The embeddings for these special tokens are generated randomly based on the dimension of the log representation used. An example is shown in Figure 1, the time elapsed for , and are set to -1. The log event-level representation and positional or temporal embedding are summed as the input feature of the transformer structure.

\ 3.5 Supervised Binary Classification Under this training objective, we utilize the output of the first token of the transformer model while ignoring the outputs of the other tokens. This output of the first token is designed to aggregate the information of the whole input log sequence, similar to the token of the BERT model, which provides an aggregated representation of the token sequence. Therefore, we consider the output of this token as a sequence-level representation. We train the model with a binary classification objective (i.e., Binary Cross Entropy Loss) with this representation.

\

:::info Authors:

  1. Xingfang Wu
  2. Heng Li
  3. Foutse Khomh

:::

:::info This paper is available on arxiv under CC by 4.0 Deed (Attribution 4.0 International) license.

:::

\

Disclaimer: The articles reposted on this site are sourced from public platforms and are provided for informational purposes only. They do not necessarily reflect the views of MEXC. All rights remain with the original authors. If you believe any content infringes on third-party rights, please contact [email protected] for removal. MEXC makes no guarantees regarding the accuracy, completeness, or timeliness of the content and is not responsible for any actions taken based on the information provided. The content does not constitute financial, legal, or other professional advice, nor should it be considered a recommendation or endorsement by MEXC.

You May Also Like

Vitalik Buterin Proposes Ethereum Gas Futures Market for Long-Term Fee Predictability

Vitalik Buterin Proposes Ethereum Gas Futures Market for Long-Term Fee Predictability

The post Vitalik Buterin Proposes Ethereum Gas Futures Market for Long-Term Fee Predictability appeared on BitcoinEthereumNews.com. Vitalik Buterin proposes an on-chain futures market for Ethereum gas, allowing users to pre-buy and lock in fees before potential price surges. This mechanism would provide long-term predictability for BASEFEE, helping developers and businesses plan transactions amid network volatility. Buterin’s vision introduces futures trading for gas, securing costs in advance for future Ethereum transactions. This system generates market-driven signals for BASEFEE evolution, reducing uncertainty in fee planning. Early projects like Oiler have tested gas derivatives, but a mature market is needed; Ethereum’s BASEFEE has fluctuated up to 200% in past cycles, per network data. Ethereum gas futures: Vitalik Buterin’s plan to pre-buy fees and stabilize costs. Discover how this on-chain market could transform transaction predictability—explore Ethereum’s future now! What is Vitalik Buterin’s Proposal for Pre-Buying Ethereum Gas? Vitalik Buterin, Ethereum’s co-founder, is advocating for an on-chain futures market that enables users to pre-buy gas at fixed prices, addressing the network’s long-standing issue of unpredictable transaction fees. This approach shifts focus from immediate cost reductions to long-term fee stability, allowing individuals and organizations to hedge against future spikes in BASEFEE. By creating a dedicated trading platform within Ethereum, Buterin aims to make gas pricing more transparent and manageable, fostering greater confidence in the ecosystem’s economic model. How Would an Ethereum Gas Futures Market Function? Ethereum’s current gas fee system relies on dynamic pricing through the EIP-1559 mechanism, where BASEFEE adjusts based on network congestion, often leading to volatility that can surge by over 150% during peak periods, as observed in historical data from the Ethereum Foundation’s reports. Buterin’s proposed futures market would operate as a decentralized exchange for gas contracts, where traders buy and sell claims to future gas units at agreed-upon prices. This market-driven mechanism would aggregate collective expectations, providing real-time signals on anticipated BASEFEE trends—such as potential increases tied…
Share
BitcoinEthereumNews2025/12/07 18:31
UK Looks to US to Adopt More Crypto-Friendly Approach

UK Looks to US to Adopt More Crypto-Friendly Approach

The post UK Looks to US to Adopt More Crypto-Friendly Approach appeared on BitcoinEthereumNews.com. The UK and US are reportedly preparing to deepen cooperation on digital assets, with Britain looking to copy the Trump administration’s crypto-friendly stance in a bid to boost innovation.  UK Chancellor Rachel Reeves and US Treasury Secretary Scott Bessent discussed on Tuesday how the two nations could strengthen their coordination on crypto, the Financial Times reported on Tuesday, citing people familiar with the matter.  The discussions also involved representatives from crypto companies, including Coinbase, Circle Internet Group and Ripple, with executives from the Bank of America, Barclays and Citi also attending, according to the report. The agreement was made “last-minute” after crypto advocacy groups urged the UK government on Thursday to adopt a more open stance toward the industry, claiming its cautious approach to the sector has left the country lagging in innovation and policy.  Source: Rachel Reeves Deal to include stablecoins, look to unlock adoption Any deal between the countries is likely to include stablecoins, the Financial Times reported, an area of crypto that US President Donald Trump made a policy priority and in which his family has significant business interests. The Financial Times reported on Monday that UK crypto advocacy groups also slammed the Bank of England’s proposal to limit individual stablecoin holdings to between 10,000 British pounds ($13,650) and 20,000 pounds ($27,300), claiming it would be difficult and expensive to implement. UK banks appear to have slowed adoption too, with around 40% of 2,000 recently surveyed crypto investors saying that their banks had either blocked or delayed a payment to a crypto provider.  Many of these actions have been linked to concerns over volatility, fraud and scams. The UK has made some progress on crypto regulation recently, proposing a framework in May that would see crypto exchanges, dealers, and agents treated similarly to traditional finance firms, with…
Share
BitcoinEthereumNews2025/09/18 02:21