Mean Reciprocal Rank

By Nick Craswell. Encyclopedia of Database Systems, 2009

316 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/c25e5d5911d2b9d3cc22601c2ef4083cbc326707

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Evaluation by Association: A Systematic Study of Quantitative Word Association Evaluation (2017, result): “…compared to all other models in our comparison. paragram-300d , (Wieting et al., 2015)) further re-ﬁned using linguistic constraints ( paragram+cf-300d , (Mrkˇsi´c et al., 2016)); (3) Multilingual embedding models from Luong et al. (2015) ( biskip-256d ) and Faruqui and Dyer (2014) ( bicca-512d ).”
TranS^3: A Transformer-based Framework for Unifying Code Summarization and Code Search (2020, methodology): “e performance of code search is evaluated in terms of four widely-used metrics: MRR (Mean Reciprocal Rank) [52], nDCG (normalized Discounted Cumulative Gain) [53] and Success Rate@k [54], where MRR measures the average reciprocal ranks of results given a set of queries and the reciprocal rank of a ”
An Introduction to Neural Information Retrieval (2018, unknown): “Mean reciprocal rank (MRR) Mean reciprocal rank [Craswell, 2009] is also computed over binary relevance judgments.”
Overview of the TREC 2021 Deep Learning Track (2025, unknown): “The reported metrics include Reciprocal Rank (RR) [Craswell, 2009], Normalized Discounted Cumulative Gains (NDCG) [Järvelin and Kekäläinen, 2002], Normalized Cumulative Gains (NCG) [Rosset et al., 2018], and Average Precision (AP) [Zhu, 2004] computed using the NIST judgments.”
An Updated Duet Model for Passage Re-ranking (2019, methodology): “The ranking model is evaluated on this passage re-ranking task using the mean reciprocal rank (MRR) metric [Craswell, 2009].”
Conformer-Kernel with Query Term Independence for Document Retrieval (2020, methodology): “We compare different runs based on following three metrics: mean reciprocal rank (MRR) [Craswell, 2009], normalized discounted cumulative gain (NDCG) [Järvelin and Kekäläinen, 2002], and normalized cumulative gain (NCG) [Rosset et al., 2018].”
Probabilistic Latent Document Network Embedding (2014, methodology): “We borrow a metric from information retrieval, called mean reciprocal rank [38] or MRR, which is defined in Equation 7, where E′ is the set of hidden links, and rank(eij) is the ranking of the hidden link eij among the unobserved links of the document from which it is hidden.”
TF-Ranking: Scalable TensorFlow Library for Learning-to-Rank (2018, methodology): “ng have been proposed in the literature. For example, widely-utilized metrics such as Normalized Discounted Cumulative Gain (NDCG) [20], Expected Reciprocal Rank (ERR) [8], Mean Reciprocal Rank (MRR) [14], Mean Average Precision (MAP), and Average Relevance Position (ARP) [48] are designed to emphas”
Multi-style Generative Reading Comprehension (2019, background): “Table 4: Passage re-ranking performance for MAP and MRR (Craswell and Robertson, 2009; Craswell, 2009) on the ANS dev.”
Document Re-Ranking Model for Machine-Reading and Comprehension (2020, methodology): “We used mean reciprocal rank at 10 (MRR@10) [24], which represents the MRR score of documents ranked in the top 10 because it is essential for relevant documents to be highly ranked for MRC models, as expressed in Equation (7):”
LLMs Meet Isolation Kernel: Lightweight, Learning-free Binary Embeddings for Fast Retrieval (2026, unknown): “• We conducted comprehensive experiments, where the main findings are: 1) In the exhaustive search , IKE generally achieves 2.5-16.7x speedup in search time and 8-16x storage reduction, while maintaining performance of MRR@10 between 98% and 101% of the original LLM space; 2) In ANN (Approximate Nea”
Chunking Is Not the Bottleneck: Why RAG Evaluation Must Look beyond Segmentation Strategy (2026, unknown): “Despite an MRR spread of 0.052 across the four main strategies (0.729–0.781), correctness varies by only 0.10 points on a 5-point scale (3.55–3.65), and faithfulness is effectively constant at 4.80–4.83.”