MIT’s new ‘recursive’ framework lets LLMs process 10 million tokens without context rot

While standard models suffer from context rot as data grows, MIT’s new Recursive Language Model (RLM) framework treats ...
IEEE

The Language of Actions: Recovering the Syntax and Semantics of Goal-Directed Human Activities

Abstract: This paper describes a framework for modeling human activities as temporally structured processes. Our approach is motivated by the inherently hierarchical nature of human activities and the ...
Frontiers

Investigating the role of the left inferior frontal gyrus in language evolution: insights from comparative neuroscience

The evolutionary adaptation of the left inferior frontal gyrus is considered a crucial neural specialization supporting the emergence of human language. As a central node in the language network, it ...
IEEE

A Recursive Decomposition Method for Large Scale Continuous Optimization

Abstract: Cooperative co-evolution (CC) is an evolutionary computation framework that can be used to solve high-dimensional optimization problems via a “divide-and-conquer” mechanism. However, the ...
Frontiers

LLMs achieve adult human performance on higher-order theory of mind tasks

We introduce a new benchmark, MoToMQA, to assess human and LLM ToM abilities at increasing orders. MoToMQA is based upon the format of the Imposing Memory Task (IMT), a well-validated psychological ...