works on

From the 1 of 19 linked papers with an AI index.

most citedGrowing a Tail: Increasing Output Diversity in Large Language Models

1 citations · 1 across the 1 of their papers we have counts for

collaborators

19 papers

cs.CL20261 cited

Growing a Tail: Increasing Output Diversity in Large Language Models

Michal Shur-Ofry, Bar Horowitz-Amsalem, Adir Rahamim +1

The paper investigates how narrowly large language models generate answers compared to the broader range of human responses, and shows that increasing temperature, using diverse pr…

cs.LG2026

Decomposing Query-Key Feature Interactions Using Contrastive Covariances

Andrew Lee, Yonatan Belinkov, Fernanda Viégas +1

Despite the central role of attention heads in Transformers, we lack tools to understand why a model attends to a particular token. To address this, we study the query-key (QK) spa…

cs.AI2026

Investigating the Development of Task-Oriented Communication in Vision-Language Models

Boaz Carmeli, Orr Paradise, Shafi Goldwasser +2

We investigate whether \emph{LLM-based agents} can develop task-oriented communication protocols that differ from standard natural language in collaborative reasoning tasks. Our fo…

cs.AI2026

CtD: Composition through Decomposition in Emergent Communication

Boaz Carmeli, Ron Meir, Yonatan Belinkov

Compositionality is a cognitive mechanism that allows humans to systematically combine known concepts in novel ways. This study demonstrates how artificial neural agents acquire an…

cs.CL2026

Will it Merge? On The Causes of Model Mergeability

Adir Rahamim, Asaf Yehudai, Boaz Carmeli +3

Model merging has emerged as a promising technique for combining multiple fine-tuned models into a single multitask model without retraining. However, the factors that determine wh…

cs.CL2025

Structured RAG for Answering Aggregative Questions

Omri Koshorek, Niv Granot, Aviv Alloni +6

Retrieval-Augmented Generation (RAG) has become the dominant approach for answering questions over large corpora. However, current datasets and methods are highly focused on cases…