paper-with-me

홈 › Papers

"Knowing When You Don't Know": A Multilingual Relevance Assessment Dataset for Robust Retrieval-Augmented Generation

2023-12-18 · Nandan Thakur, Luiz Bonifacio, Xinyu Zhang, Odunayo Ogundepo, Ehsan Kamalloo, David Alfonso-Hermelo, Xiaoguang Li, Qun Liu, Boxing Chen, Mehdi Rezagholizadeh, Jimmy Lin

Retrieval-Augmented Generation (RAG) grounds Large Language Model (LLM) output by leveraging external knowledge sources to reduce factual hallucinations. However, prior work lacks a comprehensive evaluation of different language families, making it challenging to evaluate LLM robustness against errors in external retrieved knowledge. To overcome this, we establish NoMIRACL, a human-annotated dataset for evaluating LLM robustness in RAG across 18 typologically diverse languages. NoMIRACL includes both a non-relevant and a relevant subset. Queries in the non-relevant subset contain passages judged as non-relevant, whereas queries in the relevant subset include at least a single judged relevant passage. We measure relevance assessment using: (i) hallucination rate, measuring model tendency to hallucinate, when the answer is not present in passages in the non-relevant subset, and (ii) error rate, measuring model inaccuracy to recognize relevant passages in the relevant subset.In our work, we observe that most models struggle to balance the two capacities. Models such as LLAMA-2 and Orca-2 achieve over 88% hallucination rate on the non-relevant subset. Mistral and LLAMA-3 hallucinate less but can achieve up to a 74.9% error rate on the relevant subset. Overall, GPT-4 is observed to provide the best tradeoff on both subsets, highlighting future work necessary to improve LLM robustness. NoMIRACL dataset and evaluation code are available at: https://github.com/project-miracl/nomiracl.

📄 PDF Abstract BibTeX arXiv:2312.11361

Code (1)

project-miracl/nomiracl 공식 구현

Tasks

HallucinationLanguage ModellingLarge Language ModelRAGRetrievalRetrieval-augmented Generation

Methods 이 논문이 사용한 방법론

Flan-T5 Flan-T5 is the instruction fine-tuned version of T5 or Text-to-Text Transfer Transformer Language Model.
Multi-Head Attention 설명 없음
Attention 설명 없음
WordPiece 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Position-Wise Feed-Forward Layer 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

LAMAR: An Open Language-Aware Multilingual Alignment Reranker

2026-07-24 · Seongtae Hong, Youngjoon Jang, Jungseob Lee, Seungyoon Lee 외 hf

In multilingual retrieval augmented generation, a retriever can retrieve relevant documents written in multiple languages, which are subsequently reranked before answer generation. However, it remains unclear whether exi…

Answer Generation

How Relevant is the Long Tail? A Relevance Assessment Study on Million Short

2016-06-20 · Schaer Philipp, Mayr Philipp, Sünkler Sebastian, Lewandowski Dirk

Users of web search engines are known to mostly focus on the top ranked results of the search engine result page. While many studies support this well known information seeking pattern only few studies concentrate on the…

Dealing with Unknown Variances in Best-Arm Identification

2022-10-03 · Marc Jourdan, Rémy Degenne, Emilie Kaufmann

The problem of identifying the best arm among a collection of items having Gaussian rewards distribution is well understood when the variances are known. Despite its practical relevance for many applications, few works s…

Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG

2025-09-17 · Dayeon Ki, Marine Carpuat, Paul McNamee, Daniel Khashabi 외 arxiv

Multilingual Retrieval-Augmented Generation (mRAG) systems enable language models to answer knowledge-intensive queries with citation-supported responses across languages. Despite their growing use, an open questions is …

Efficient Medical Image Assessment via Self-supervised Learning

2022-09-28 · Chun-Yin Huang, Qi Lei, Xiaoxiao Li

High-performance deep learning methods typically rely on large annotated training datasets, which are difficult to obtain in many clinical applications due to the high cost of medical image labeling. Existing data assess…

Self-Supervised Learning