mlsanityData to 7 Oct 2026

LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion

arXiv 2306.02561 ↗First posted 5 Jun 2023cs.CL cs.AI cs.LGcited by 550

Not on the map. Why?

Cited by 111 in LLM alignment / RLHF / preference optimisation, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We present LLM-Blender, an ensembling framework designed to attain consistently superior performance by leveraging the diverse strengths of multiple open-source large language models (LLMs). Our framework consists of two modules: PairRanker and GenFuser, addressing the observation that optimal LLMs for different examples can significantly vary. PairRanker employs a specialized pairwise comparison method to distinguish subtle differences between candidate outputs. It jointly encodes the input text and a pair of candidates, using cross-attention encoders to determine the superior one. Our results demonstrate that PairRanker exhibits the highest correlation with ChatGPT-based ranking. Then, GenFuser aims to merge the top-ranked candidates, generating an improved output by capitalizing on their strengths and mitigating their weaknesses. To facilitate large-scale evaluation, we introduce a benchmark dataset, MixInstruct, which is a mixture of multiple instruction datasets featuring oracle pairwise comparisons. Our LLM-Blender significantly outperform individual LLMs and baseline methods across various metrics, establishing a substantial performance gap.

Key facts. Cited by 550 papers in the map (data through 7 Oct 2026). Builds on 26 of its 53 references that are in the map. First posted on arXiv 5 Jun 2023.

Builds on

26 of 53 references.

References · 26

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  5. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  6. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  9. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  10. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  11. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
  12. 2023Pythia: A Suite for Analyzing Large Language Models Across Training and Scalingcited by 1,734
14 more
  1. 2021DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharingcited by 1,219
  2. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
  3. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,196
  4. 2019PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarizationcited by 1,100
  5. 2015Teaching Machines to Read and Comprehendcited by 1,039
  6. 2021GLM: General Language Model Pretraining with Autoregressive Blank Infillingcited by 1,026
  7. 2018Adafactor: Adaptive Learning Rates with Sublinear Memory Costcited by 896
  8. 2016Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyondcited by 809
  9. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 738
  10. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 719
  11. 2021SimCLS: A Simple Framework for Contrastive Learning of Abstractive Summarizationcited by 131
  12. 2022SummaReranker: A Multi-Task Mixture-of-Experts Re-ranking Framework for Abstractive Summarizationcited by 61
  13. 2022Towards Summary Candidates Fusioncited by 8
  14. 2019Ensemble approach for natural language question answering problemcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 18; 2024: 142; 2025: 226; 2026: 164.

550 papers.

Sort:

By area:

  1. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 951
  2. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  3. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
  4. 2024Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Modelscited by 431
  5. 2025Multi-Agent Collaboration Mechanisms: A Survey of LLMscited by 424
  6. 2024Mixture-of-Agents Enhances Large Language Model Capabilitiescited by 420
  7. 2024Hybrid LLM: Cost-Efficient and Quality-Aware Query Routingcited by 354
  8. 2024Skywork-Reward: Bag of Tricks for Reward Modeling in LLMscited by 322
  9. 2023ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMscited by 297
  10. 2024LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Taskscited by 236
  11. 2024Self-Play Preference Optimization for Language Model Alignmentcited by 228
  12. 2024RouterBench: A Benchmark for Multi-LLM Routing Systemcited by 222
38 more
  1. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 215
  2. 2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 198
  3. 2024Advancing LLM Reasoning Generalists with Preference Treescited by 191
  4. 2023Mitigating the Alignment Tax of RLHFcited by 181
  5. 2024More Agents Is All You Needcited by 174
  6. 2023Large Language Model Routing with Benchmark Datasetscited by 166
  7. 2025Multi-agent Architecture Search via Agentic Supernetcited by 164
  8. 2024Regularizing Hidden States Enables Learning Generalizable Reward Model for LLMscited by 142
  9. 2023Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Modelscited by 141
  10. 2024Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systemscited by 129
  11. 2024Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffoldingcited by 115
  12. 2024The Good, The Bad, and The Greedy: Evaluation of LLMs Should Not Ignore Non-Determinismcited by 113
  13. 2024Knowledge Fusion of Large Language Modelscited by 110
  14. 2024G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networkscited by 110
  15. 2024Critique-out-Loud Reward Modelscited by 105
  16. 2024Aligner: Efficient Alignment by Learning to Correctcited by 96
  17. 2024Decoding-Time Language Model Alignment with Multiple Objectivescited by 90
  18. 2023Deep Model Fusion: A Surveycited by 89
  19. 2024β-DPO: Direct Preference Optimization with Dynamic βcited by 89
  20. 2025Harnessing Multiple Large Language Models: A Survey on LLM Ensemblecited by 86
  21. 2024RouterDC: Query-Based Router by Dual Contrastive Learning for Assembling Large Language Modelscited by 84
  22. 2025Universal Model Routing for Efficient LLM Inferencecited by 80
  23. 2024RRM: Robust Reward Model Training Mitigates Reward Hackingcited by 75
  24. 2024Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHFcited by 73
  25. 2023TIGERScore: Towards Building Explainable Metric for All Text Generation Taskscited by 69
  26. 2024Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaborationcited by 66
  27. 2024Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimizationcited by 66
  28. 2024LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Playcited by 65
  29. 2024Instructional Fingerprinting of Large Language Modelscited by 64
  30. 2024MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?cited by 64
  31. 2024Self-Exploring Language Models: Active Preference Elicitation for Online Alignmentcited by 63
  32. 2025BEST-Route: Adaptive LLM Routing with Test-Time Optimal Computecited by 63
  33. 2025LVAgent: Long Video Understanding by Multi-Round Dynamical Collaboration of MLLM Agentscited by 58
  34. 2025RouterEval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMscited by 58
  35. 2025FlowReasoner: Reinforcing Query-Level Meta-Agentscited by 56
  36. 2025Atom of Thoughts for Markov LLM Test-Time Scalingcited by 54
  37. 2024WPO: Enhancing RLHF with Weighted Preference Optimizationcited by 53
  38. 2024FuseChat: Knowledge Fusion of Chat Modelscited by 53

50 most cited of 550 All 550 (JSON)