Pith. sign in

REVIEW 25 cited by

Gecko: Versatile Text Embeddings Distilled from Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.20327 v1 pith:KETIZPUA submitted 2024-03-29 cs.CL cs.AI

classification cs.CLcs.AI
keywords geckoembeddingmodelstextachievesdatadimensionsembeddings
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present Gecko, a compact and versatile text embedding model. Gecko achieves strong retrieval performance by leveraging a key idea: distilling knowledge from large language models (LLMs) into a retriever. Our two-step distillation process begins with generating diverse, synthetic paired data using an LLM. Next, we further refine the data quality by retrieving a set of candidate passages for each query, and relabeling the positive and hard negative passages using the same LLM. The effectiveness of our approach is demonstrated by the compactness of the Gecko. On the Massive Text Embedding Benchmark (MTEB), Gecko with 256 embedding dimensions outperforms all existing entries with 768 embedding size. Gecko with 768 embedding dimensions achieves an average score of 66.31, competing with 7x larger models and 5x higher dimensional embeddings.

Discussion (0). Sign in to comment.

Forward citations

Cited by 25 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. mamabench and mamaretrieval: Benchmarks for Evaluating Medical Retrieval-Augmented Generation in Maternal, Neonatal, and Reproductive Health

    cs.CL 2026-06 unverdicted novelty 7.0 of 10

    The authors introduce and release two new benchmarks for maternal-health RAG evaluation, built from expert sources with graded labels and disclosed limitations rather than binary judgments or new question authoring.

  2. LLM2Vec-Gen: Generative Embeddings from Large Language Models

    cs.CL 2026-03 conditional novelty 7.0 of 10

    LLM2Vec-Gen learns embeddings that encode an LLM's likely response to a query, improving self-supervised MTEB scores, safety, and reasoning-intensive retrieval.

  3. AI for Statutory Simplification: A Comprehensive State Legal Corpus and Labor Benchmark

    cs.IR 2025-08 conditional novelty 7.0 of 10

    State-of-the-art LLMs with retrieval answer simplified boolean questions about state unemployment insurance law with at best 0.69 F1, well short of reliable end-to-end code simplification.

  4. BitNet Text Embeddings

    cs.CL 2026-06 unverdicted novelty 6.0 of 10

    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.

  5. Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation

    cs.LG 2026-04 unverdicted novelty 6.0 of 10

    RISE applies CountSketch to dual lexical and semantic channels derived from output-layer gradient outer products, cutting data attribution storage by up to 112x and enabling retrospective and prospective influence ana...

  6. LLMs Meet Isolation Kernel: Lightweight, Learning-free Binary Embeddings for Fast Retrieval

    cs.IR 2026-01 unverdicted novelty 6.0 of 10

    Isolation-kernel binary hashing (IKE) compresses LLM embeddings to a few hundred bytes per point with retrieval accuracy near the original and large speedups in exhaustive and ANN search.

  7. LLM-guided Hierarchical Search for End-to-end Reasoning Intensive Retrieval

    cs.IR 2025-10 conditional novelty 6.0 of 10

    An LLM directly traverses a hierarchical semantic index of a corpus, using calibrated path-relevance scores to retrieve documents for reasoning-intensive queries.

  8. Efficient Item ID Generation for Large-Scale LLM-based Recommendation

    cs.IR 2025-09 conditional novelty 6.0 of 10

    LLM-based recommenders can treat item IDs as single direct embeddings and decode in one step, with a two-level softmax for efficiency and quality matching or beating multi-token models.

  9. Negative Matters: Multi-Granularity Hard-Negative Synthesis and Anchor-Token-Aware Pooling for Enhanced Text Embeddings

    cs.CL 2025-08 conditional novelty 6.0 of 10

    A new MTEB state-of-the-art for text embeddings is reported by combining multi-granularity LLM-generated hard negatives with curriculum training and an anchor-token-aware pooling method.

  10. BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent

    cs.CL 2025-08 conditional novelty 6.0 of 10

    BrowseComp-Plus is a fixed, human-verified document corpus with supporting and negative documents that allows controlled, reproducible evaluation of deep-research agents and their retrievers.

  11. Towards Compute-Optimal Many-Shot In-Context Learning

    cs.CL 2025-07 conditional novelty 6.0 of 10

    Hybrid demonstration selection that adds 20 similar examples to a large cached random or k-means set matches or beats similarity-only selection at up to 10x lower estimated inference cost in many-shot ICL.

  12. Clustering and Median Aggregation Improve Differentially Private Inference

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Clustering seed texts and privately aggregating median token logits improves representativeness and reduces reported privacy cost for DP synthetic text generation.

  13. Redundancy, Isotropy, and Intrinsic Dimensionality of Prompt-based Text Embeddings

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Prompt-based text embeddings can be truncated to a small fraction of their dimensions with little performance loss on classification and clustering, but retrieval and STS degrade faster; the difference tracks lower in...

  14. Aligning Web Query Generation with Ranking Objectives via Direct Preference Optimization

    cs.IR 2025-05 conditional novelty 6.0 of 10

    Using direct preference optimization with reranker or GPT-3.5 preferences to align synthetic query generation improves downstream dense retrieval effectiveness on MS MARCO and TREC-DL.

  15. RaDeR: Reasoning-aware Dense Retrieval Models

    cs.CL 2025-05 conditional novelty 6.0 of 10

    A math-trained dense retriever and reranker, built from MCTS reasoning trajectories and self-reflection, outperforms strong baselines on reasoning-intensive retrieval benchmarks and beats BM25 on chain-of-thought queries.

  16. SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts

    cs.CL 2025-07 conditional novelty 5.0 of 10

    Using multiple LLM-generated summaries of the same abstract as positive pairs trains scientific text embeddings that beat citation-trained baselines on retrieval and clustering, while the new benchmark shares its trai...

  17. Towards Better Instruction Following Retrieval Models

    cs.CL 2025-05 conditional novelty 5.0 of 10

    A new training corpus and embedding model improve instruction-following p-MRR by up to 9 points on FollowIR, MAIR, and Bright benchmarks.

  18. Synthesizing and Adapting Error Correction Data for Mobile Large Language Model Applications

    cs.LG 2025-05 conditional novelty 5.0 of 10

    A production mobile-keyboard error-correction system that synthesizes large LLM-generated training data, reweights it with a differentially private federated small LM, and fine-tunes a billion-parameter LLM via LoRA r...

  19. MIRB: Mathematical Information Retrieval Benchmark

    cs.IR 2025-05 conditional novelty 5.0 of 10

    MIRB, a unified benchmark of four math retrieval tasks across 12 datasets, shows current retrieval models score far lower on premise retrieval than on semantic retrieval, and cross-encoder rerankers often hurt.

  20. Generative Data Refinement: Just Ask for Better Data

    cs.LG 2025-09 conditional novelty 4.0 of 10

    A pretrained LLM can rewrite individual data samples to strip out PII or toxic content while preserving useful information, creating safer training data.

  21. Exploring Reasoning-Infused Text Embedding with Large Language Models for Zero-Shot Dense Retrieval

    cs.CL 2025-08 conditional novelty 4.0 of 10

    Reasoning-infused text embedding, which prepends LLM-generated reasoning to queries before embedding, improves zero-shot dense retrieval on BRIGHT.

  22. QZhou-Embedding Technical Report

    cs.CL 2025-08 conditional novelty 4.0 of 10

    QZhou-Embedding reports state-of-the-art average scores on MTEB and CMTEB as of August 27, 2025, using a two-stage multi-task pipeline with LLM-based data synthesis.

  23. LGAI-EMBEDDING-Preview Technical Report

    cs.CL 2025-06 conditional novelty 4.0 of 10

    A Mistral-7B embedding model trained with in-context instructions, soft labels from an in-house retrieval pipeline, and margin-based hard-negative mining reports top-tier MTEB English v2 scores.

  24. GATE: General Arabic Text Embedding for Enhanced Semantic Textual Similarity with Matryoshka Representation Learning and Hybrid Loss Training

    cs.CL 2025-05 reject novelty 4.0 of 10

    GATE's Arabic-Triplet-Matryoshka-V2 reports the highest average scores on the MTEB Arabic STS17/STS22/STS22-v2 tasks among the models compared in the paper.

  25. A Practical Guide for Evaluating LLMs and LLM-Reliant Systems

    cs.AI 2025-06 conditional novelty 3.0 of 10

    A guide that organizes LLM evaluation into three pillars (datasets, metrics, and methodology) and introduces a '5 D's' checklist for building evaluation datasets.

Pith tools