A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Zhan-Lun Chang, Dong-Jun Han, Seyyedali Hosseinalipour, et al.
A two-stage multimodal RAG framework that separates multimodal query translation from generator-aligned reranking to bridge the semantic-utility gap.
Micah Hodosh, Peter Young, Julia Hockenmaier
Frames image description as a ranking task, introducing a benchmark of 8,000 images with five captions each and showing that ranking-based evaluation can be automated.
Unknown
RANK1 is the first reranking model trained to exploit test-time compute, enabling a new paradigm for information retrieval.
Unknown
RankRAG unifies context ranking with retrieval-augmented generation in LLMs by jointly fine-tuning a single model for both retrieval and answer generation, improving performance on knowledge-intensive tasks.
Thomas Philip Rúnarsson, Xin Yao
Introduces stochastic ranking to balance objective and penalty functions in constrained evolutionary optimization, improving search performance without specialized variation operators.
Unknown
Jina Reranker is a neural reranking model that improves search and RAG systems by reordering retrieved documents for better query alignment.
Kailin Jiang, Lei Liu, Jian Xi, et al.
Proposes a rubric-oriented framework for evaluating, diagnosing, and optimizing document set quality beyond independent scoring, achieving best downstream generation with fewer documents.