Notable data and AI papers

Papers627Labs45

627 papers · 45 labs

LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations research paper by MongoDB, 2025

MongoDB · Retrieval and data · Sep 16, 2025 · 9 citations

Read on arXiv

What it shows

LEAF, a knowledge distillation framework for text embeddings, enables flexible asymmetric architectures and robustness, achieving state-of-the-art performance in information retrieval and multi-task settings with minimal training requirements.

By Robin Vujanic, Thomas Rueckstiess · arXiv 2509.12539 · PDF

Hugging Face's summary; not yet checked by hand.

More from MongoDB

All 2
ArchitecturesDec 23, 2024ORIGAMI: A generative transformer architecture for predictions from semi-structured dataA transformer that learns directly from nested JSON documents, treating prediction as next-token generation, and stays competitive on tabular benchmarks.MongoDB2 citationsRead →

Other retrieval and data papers

Topic
Retrieval and dataUnverifiedNov 9, 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language ModelBLOOM, a 176B-parameter open-access language model, achieves competitive performance on various benchmarks and is released publicly with a Responsible AI License.Hugging Face3,165 citations41 upvotesRead →Retrieval and dataApr 24, 2024From Local to Global: A Graph RAG Approach to Query-Focused SummarizationGraphRAG builds a knowledge graph of a document set so a model can answer questions about the whole collection.Microsoft2,561 citations7 upvotesRead →Retrieval and dataUnverifiedAug 13, 2025DINOv3DINOv3, a self-supervised learning model, achieves superior performance across various vision tasks by scaling datasets and models, addressing dense feature degradation, and enhancing flexibility with post-hoc strategies.Meta1,932 citations320 upvotesRead →Retrieval and dataUnverifiedJan 8, 2026Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and RankingThe Qwen3-VL-Embedding and Qwen3-VL-Reranker models form an end-to-end multimodal search pipeline, leveraging multi-stage training and cross-attention mechanisms to achieve high-precision retrieval across diverse modalities.Alibaba (Qwen)315 citations59 upvotesRead →Retrieval and dataUnverifiedMay 23, 2024DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic DataGenerating extensive Lean 4 proof data from mathematical competition problems improved DeepSeekMath 7B's theorem-proving capabilities over GPT-4 and other methods.DeepSeek284 citations47 upvotesRead →Retrieval and dataFeb 9, 2024Aya Dataset: An Open-Access Collection for Multilingual Instruction TuningThe initiative builds a human-curated instruction-following dataset spanning 65 languages and creates the largest multilingual collection of instruction-following instances through templating and translating existing datasets across 114 languages, contributing datasets and platforms for participatory research.Cohere253 citations57 upvotesRead →
About this paper
Venue
Annual Meeting of the Association for Computational Linguistics
Citations
9, 1 influential · Semantic Scholar
Upvotes
0 · Hugging Face
Lab
MongoDB · on Companies · on Quarterly · on Paydays · on Releases · on TechConf

Changes

CitationsNov 2, 2026Citations rose from 7 to 9.
CitationsOct 5, 2026Citations rose from 6 to 7.
CitationsSep 28, 2026A first count of 1 influential citations.
CitationsSep 28, 2026A first count of 6 citations.
New paperSep 26, 2026Added to the list.