paper-with-me

Papers

Structure and Scale in Simplicial Sequence Modelling

2026-05-31 · Matthew Farrugia-Roberts arxiv

Modern large-scale deep learning exhibits two striking empirical phenomena: behavioural scaling laws (predictable performance gains with increasing scale) and emergent mechanisms (structured internal representations and circuits in deep neural networks). We hypothesise that these two phenomena are connected: that predictable changes in behaviour are the result of predictable changes in internal computational structure. In this paper, we report preliminary evidence of such a connection. We find a correlation between scaling patterns in performance and representations in small transformers trained to predict the outputs of a hidden Markov model, for which residual activations are known to linearly encode a belief distribution over latent states in a probability simplex.

📄 PDF Abstract BibTeX arXiv:2606.01302

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Applying language models to algebraic topology: generating simplicial cycles using multi-labeling in Wu's formula

2023-06-01 · Kirill Brilliantov, Fedor Pavutnitskiy, Dmitry Pasechnyuk, German Magai

Computing homotopy groups of spheres has long been a fundamental objective in algebraic topology. Various theoretical and algorithmic approaches have been developed to tackle this problem. In this paper we take a step to…

Language Modelling

Simplicial Attention Networks

2022-04-20 · Christopher Wei Jin Goh, Cristian Bodnar, Pietro Liò

Graph representation learning methods have mostly been limited to the modelling of node-wise interactions. Recently, there has been an increased interest in understanding how higher-order structures can be utilised to fu…

Graph Representation LearningRepresentation Learning

Spectral Convergence of Complexon Shift Operators

2023-09-12 · Purui Zhang, Xingchao Jian, Feng Ji, Wee Peng Tay 외

Topological Signal Processing (TSP) utilizes simplicial complexes to model structures with higher order than vertices and edges. In this paper, we study the transferability of TSP via a generalized higher-order version o…

Topological Deep Learning with State-Space Models: A Mamba Approach for Simplicial Complexes

2024-09-18 · Marco Montagna, Simone Scardapane, Lev Telyatnikov

Graph Neural Networks based on the message-passing (MP) mechanism are a dominant approach for handling graph-structured data. However, they are inherently limited to modeling only pairwise interactions, making it difficu…

MambaState Space Models

Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents

2025-10-15 · Johan Obando-Ceron, Walter Mayor, Samuel Lavoie, Scott Fujimoto 외 arxiv

Recent works have proposed accelerating the wall-clock training time of actor-critic methods via the use of large-scale environment parallelization; unfortunately, these can sometimes still require large number of enviro…

Reinforcement Learning