Behavior Structformer: Learning Players Representations with Structured Tokenization

Oleg Smirnov, Labinot Polisi
6/7/2024
cs.CLcs.LG

No confident code implementation found for this paper.

Abstract

In this paper, we introduce the Behavior Structformer, a method for modeling user behavior using structured tokenization within a Transformer-based architecture. By converting tracking events into dense tokens, this approach enhances model training efficiency and effectiveness. We demonstrate its superior performance through ablation studies and benchmarking against traditional tabular and semi-structured baselines. The results indicate that structured tokenization with sequential processing significantly improves behavior modeling.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

player2vec: A Language Modeling Approach to Understand Player Behavior in Games

83% match2024

Task Tokens: A Flexible Approach to Adapting Behavior Foundation Models

80% match2025

Learning to Play Atari in a World of Tokens

79% match1 repo112024

PANTHER: Generative Pretraining Beyond Language for Sequential User Behavior Modeling

78% match1 repo132025

TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters

78% match2 repos5922024

Behavior-Equivalent Token: Single-Token Replacement for Long Prompts in LLMs

77% match2025

Cite this paper

@article{smirnov2024behavior,
  title  = {Behavior Structformer: Learning Players Representations with Structured Tokenization},
  author = {Oleg Smirnov and Labinot Polisi},
  year   = {2024},
  eprint = {2406.05274},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.05274v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.05274.svg)](https://codeofpaper.com/paper/2406.05274)

Discussion