Proves Voronoi complexity equals sign-rank for top-1 retrieval, introduces CUS diagnostic predicting retrieval failure at AUC >0.8 without labels, and AT-DW-InfoNCE objective with derived alpha^*=2.0 that improves Recall@100 on synthetic data.
Gecko: Versatile text embeddings distilled from large language models.arXiv preprint arXiv:2403.20327, 2024
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
representative citing papers
IdioLink introduces a benchmark dataset and evaluation showing that strong embedding models struggle to retrieve equivalent meanings across idiomatic and literal forms, relying on shallow cues instead.
mamabench and mamaretrieval release scope-filtered QA and graded chunk-level relevance benchmarks for maternal, neonatal, and reproductive-health RAG, with disclosed label limits.
BITEMBED trains 1.58-bit ternary-weight LLM embedders with contrastive pre-training, supervised distillation, and multi-precision output training, matching FP16 teachers within ~0.6 MMTEB points at ~2x CPU speed.
MM-Matryoshka is a 2D Matryoshka training framework enabling budget-elastic ColPali-style multi-vector visual document retrieval along dimension and layer without separate models per budget.
A survey that categorizes RIR benchmarks by domain and modality, proposes a taxonomy for integrating reasoning into retrieval pipelines, and outlines key challenges.
LLMs exhibit mid-layer representation advantage for recommendations; MARC compresses representations modularly to reduce costs while improving performance, as shown in a large-scale online advertising deployment.
Influence scoring can use only forward passes: CountSketch-compressed outer products of the LM-head residual and final hidden state give accurate attribution and valuation from 14M to 32B parameters.
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
NV-Embed achieves first place on the MTEB leaderboard across 56 tasks by combining a latent attention layer, causal-mask removal, two-stage contrastive training, and data curation for LLM-based embedding models.
PETRA is a curated 1.36M-chunk petroleum-engineering retrieval dataset and pipeline that raises in-domain nDCG from 0.703 to 0.763 via score fusion and delivers 44% relative gain on an Earth Science benchmark through reranker adaptation on synthetic supervision.
Traits Run Deeper proposes MFR, TSMF asymmetric fusion, and DCPR modules to improve multimodal personality assessment, claiming 25% MSE reduction and first place on AVI Challenge 2026.
Lightweight proxy models deliver over 100x cost and latency savings for semantic AI queries in databases with accuracy preserved or improved on benchmarks up to 10M rows.
A 300M multilingual embedding model matches or exceeds 7B retrieval performance via optimized data scale, hard negatives, and task diversity over language diversity.
A native multimodal embedding model from Gemini achieves reported state-of-the-art results on retrieval benchmarks across modalities via large-scale contrastive learning.
The paper runs a systematic comparison of RAG pipelines, retrieval strategies, embedding models, and LLMs on space-domain documents and reports gains in accuracy, relevance, and decision support.
citing papers explorer
-
The Voronoi Bottleneck: Capacity-Aware Dense Retrieval for Product Search
Proves Voronoi complexity equals sign-rank for top-1 retrieval, introduces CUS diagnostic predicting retrieval failure at AUC >0.8 without labels, and AT-DW-InfoNCE objective with derived alpha^*=2.0 that improves Recall@100 on synthetic data.
-
IdioLink: Retrieving Meaning Beyond Words Across Idiomatic and Literal Expressions
IdioLink introduces a benchmark dataset and evaluation showing that strong embedding models struggle to retrieve equivalent meanings across idiomatic and literal forms, relying on shallow cues instead.
-
mamabench and mamaretrieval: Benchmarks for Evaluating Medical Retrieval-Augmented Generation in Maternal, Neonatal, and Reproductive Health
mamabench and mamaretrieval release scope-filtered QA and graded chunk-level relevance benchmarks for maternal, neonatal, and reproductive-health RAG, with disclosed label limits.
-
BitNet Text Embeddings
BITEMBED trains 1.58-bit ternary-weight LLM embedders with contrastive pre-training, supervised distillation, and multi-precision output training, matching FP16 teachers within ~0.6 MMTEB points at ~2x CPU speed.
-
MM-Matryoshka: Towards Budget-Elastic Visual Document Retrieval via a 2D Multimodal Matryoshka Training Framework
MM-Matryoshka is a 2D Matryoshka training framework enabling budget-elastic ColPali-style multi-vector visual document retrieval along dimension and layer without separate models per budget.
-
A Survey of Reasoning-Intensive Retrieval: Progress and Challenges
A survey that categorizes RIR benchmarks by domain and modality, proposes a taxonomy for integrating reasoning into retrieval pipelines, and outlines key challenges.
-
Modular Representation Compression: Adapting LLMs for Efficient and Effective Recommendations
LLMs exhibit mid-layer representation advantage for recommendations; MARC compresses representations modularly to reduce costs while improving performance, as shown in a large-scale online advertising deployment.
-
Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation
Influence scoring can use only forward passes: CountSketch-compressed outer products of the LM-head residual and final hidden state give accurate attribution and valuation from 14M to 32B parameters.
-
ARHN: Answer-Centric Relabeling of Hard Negatives with Open-Source LLMs for Dense Retrieval
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
-
NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models
NV-Embed achieves first place on the MTEB leaderboard across 56 tasks by combining a latent attention layer, causal-mask removal, two-stage contrastive training, and data curation for LLM-based embedding models.
-
PETRA: Transforming Web Text for Petroleum-Engineering Domain Adaptation
PETRA is a curated 1.36M-chunk petroleum-engineering retrieval dataset and pipeline that raises in-domain nDCG from 0.703 to 0.763 via score fusion and delivers 44% relative gain on an Earth Science benchmark through reranker adaptation on synthetic supervision.
-
Traits Run Deeper: Trait-Specific Asymmetric Fusion for Personality Assessment
Traits Run Deeper proposes MFR, TSMF asymmetric fusion, and DCPR modules to improve multimodal personality assessment, claiming 25% MSE reduction and first place on AVI Challenge 2026.
-
100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models
Lightweight proxy models deliver over 100x cost and latency savings for semantic AI queries in databases with accuracy preserved or improved on benchmarks up to 10M rows.
-
Retrofitting Small Multilingual Models for Retrieval: Matching 7B Performance with 300M Parameters
A 300M multilingual embedding model matches or exceeds 7B retrieval performance via optimized data scale, hard negatives, and task diversity over language diversity.
-
Gemini Embedding 2: A Native Multimodal Embedding Model from Gemini
A native multimodal embedding model from Gemini achieves reported state-of-the-art results on retrieval benchmarks across modalities via large-scale contrastive learning.
-
A Systematic Evaluation of Retrieval-Augmented Generation and Language Models for Space Operations
The paper runs a systematic comparison of RAG pipelines, retrieval strategies, embedding models, and LLMs on space-domain documents and reports gains in accuracy, relevance, and decision support.