5 papers
Optimal Budgeted Adaptation of Large Language Models
Jing Wang, Jie Shen, Dean Foster +2
The trade-off between labeled data availability and downstream accuracy remains a central challenge in fine-tuning large language models (LLMs). We propose a principled framework f…
Linguistic and Argument Diversity in Synthetic Data for Function-Calling Agents
Dan Greenstein, Zohar Karnin, Chen Amiraz +1
The construction of function calling agents has emerged as a promising avenue for extending model capabilities. A major challenge for this task is obtaining high quality diverse da…
The Cross-Lingual Cost: Retrieval Biases in RAG over Arabic-English Corpora
Chen Amiraz, Yaroslav Fyodorov, Elad Haramaty +2
Cross-lingual retrieval-augmented generation (RAG) is a critical capability for retrieving and generating answers across languages. Prior work in this context has mostly focused on…
The Distracting Effect: Understanding Irrelevant Passages in RAG
Chen Amiraz, Florin Cuconasu, Simone Filice +1
A well-known issue with Retrieval Augmented Generation (RAG) is that retrieved passages that are irrelevant to the query sometimes distract the answer-generating LLM, causing it to…
Generating Diverse Q&A Benchmarks for RAG Evaluation with DataMorgana
Simone Filice, Guy Horowitz, David Carmel +3
Evaluating Retrieval-Augmented Generation (RAG) systems, especially in domain-specific contexts, requires benchmarks that address the distinctive requirements of the applicative sc…