paper-with-me

Papers

Mathematically Modeling the Lexicon Entropy of Emergent Language

2022-11-28 · Brendon Boldt, David Mortensen

We formulate a stochastic process, FiLex, as a mathematical model of lexicon entropy in deep learning-based emergent language systems. Defining a model mathematically allows it to generate clear predictions which can be directly and decisively tested. We empirically verify across four different environments that FiLex predicts the correct correlation between hyperparameters (training steps, lexicon size, learning rate, rollout buffer size, and Gumbel-Softmax temperature) and the emergent language's entropy in 20 out of 20 environment-hyperparameter combinations. Furthermore, our experiments reveal that different environments show diverse relationships between their hyperparameters and entropy which demonstrates the need for a model which can make well-defined predictions at a precise level of granularity.

📄 PDF Abstract BibTeX arXiv:2211.15783

Code (1)

brendon-boldt/filex-emergent-language 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Modeling Emergent Lexicon Formation with a Self-Reinforcing Stochastic Process

2022-06-22 · Brendon Boldt, David Mortensen

We introduce FiLex, a self-reinforcing stochastic process which models finite lexicons in emergent language experiments. The central property of FiLex is that it is a self-reinforcing process, parallel to the intuition t…

LexMAE: Lexicon-Bottlenecked Pretraining for Large-Scale Retrieval

2022-08-31 · Tao Shen, Xiubo Geng, Chongyang Tao, Can Xu 외

In large-scale retrieval, the lexicon-weighting paradigm, learning weighted sparse representations in vocabulary space, has shown promising results with high quality and low latency. Despite it deeply exploiting the lexi…

CPUDecoderLanguage ModelingLanguage Modelling+2

Searching for the Most Human-like Emergent Language

2025-10-03 · Brendon Boldt, David Mortensen arxiv

In this paper, we design a signalling game-based emergent communication environment to generate state-of-the-art emergent languages in terms of similarity to human language. This is done with hyperparameter optimization,…

Hyperparameter OptimizationTransfer Learning

Large language models can disambiguate opioid slang on social media

2026-03-11 · Kristy A. Carpenter, Issah A. Samori, Mathew V. Kiang, Keith Humphreys 외 arxiv

Social media text shows promise for monitoring trends in the opioid overdose crisis; however, the overwhelming majority of social media text is unrelated to opioids. When leveraging social media text to monitor trends in…

Apparent Communicative Efficiency in the Lexicon is Emergent

2021-02-01 · SCiL 2021 2 · Spencer Caplan, Jordan Kodner, Charles Yang