paper-with-me

Papers

Linear Latent World Models in Simple Transformers: A Case Study on Othello-GPT

2023-10-11 · Dean S. Hazineh, Zechen Zhang, Jeffery Chiu

Foundation models exhibit significant capabilities in decision-making and logical deductions. Nonetheless, a continuing discourse persists regarding their genuine understanding of the world as opposed to mere stochastic mimicry. This paper meticulously examines a simple transformer trained for Othello, extending prior research to enhance comprehension of the emergent world model of Othello-GPT. The investigation reveals that Othello-GPT encapsulates a linear representation of opposing pieces, a factor that causally steers its decision-making process. This paper further elucidates the interplay between the linear world representation and causal decision-making, and their dependence on layer depth and model complexity. We have made the code public.

📄 PDF Abstract BibTeX arXiv:2310.07582

Code (2)

deanhazineh/emergent-world-representations-othello 공식 구현 pytorch
alxndrtl/othello_mamba pytorch

Tasks

Decision Making

Similar Papers 제목 키워드 기반

Why Attention Fails: The Degeneration of Transformers into MLPs in Time Series Forecasting

2025-09-25 · Zida Liang, Jiayi Zhu, Weiqiang Sun arxiv

Transformer-based architectures achieved high performance in natural language processing and computer vision, yet many studies have shown that they have not demonstrated a clear advantage in time series forecasting and e…

Time Series Forecasting

Next-Latent Prediction Transformers Learn Compact World Models

2025-11-08 · Jayden Teoh, Manan Tomar, Kwangjun Ahn, Edward S. Hu 외 arxiv

Transformers replace recurrence with a memory that grows with sequence length and self-attention that enables ad-hoc lookups over past tokens. Consequently, they lack an inherent incentive to compress history into compac…

JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention

2023-10-01 · Yuandong Tian, Yiping Wang, Zhenyu Zhang, Beidi Chen 외

We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures. This is achieved by integrating out the self-attention layer in…

Latte: Latent Attention for Linear Time Transformers

2024-02-27 · Rares Dolga, Lucas Maystre, Marius Cobzarenco, David Barber

The time complexity of the standard attention mechanism in transformers scales quadratically with sequence length. We propose a probabilistic framework for attention, enabling us to derive a novel low-rank linear re-para…

Text Generation

GIVT: Generative Infinite-Vocabulary Transformers

2023-12-04 · Michael Tschannen, Cian Eastwood, Fabian Mentzer

We introduce Generative Infinite-Vocabulary Transformers (GIVT) which generate vector sequences with real-valued entries, instead of discrete tokens from a finite vocabulary. To this end, we propose two surprisingly simp…

Conditional Image GenerationDecoderDepth EstimationImage Generation+1