mlsanityData to 7 Oct 2026

NEFTune: Noisy Embeddings Improve Instruction Finetuning

arXiv 2310.05914 ↗First posted 9 Oct 2023cs.CL cs.LGcited by 99

Not on the map. Why?

Abstract (arXiv)

We show that language model finetuning can be improved, sometimes dramatically, with a simple augmentation. NEFTune adds noise to the embedding vectors during training. Standard finetuning of LLaMA-2-7B using Alpaca achieves 29.79% on AlpacaEval, which rises to 64.69% using noisy embeddings. NEFTune also improves over strong baselines on modern instruction datasets. Models trained with Evol-Instruct see a 10% improvement, with ShareGPT an 8% improvement, and with OpenPlatypus an 8% improvement. Even powerful models further refined with RLHF such as LLaMA-2-Chat benefit from additional training with NEFTune.

Key facts. Cited by 99 papers in the map (data through 7 Oct 2026). Builds on 23 of its 37 references that are in the map. First posted on arXiv 9 Oct 2023.

Builds on

23 of 37 references.

References · 23

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  6. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  7. 2019HellaSwag: Can a Machine Really Finish Your Sentence?cited by 4,018
  8. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  9. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  10. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,355
  11. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  12. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
11 more
  1. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  2. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,462
  3. 2023WizardLM: Empowering large pre-trained language models to follow complex instructionscited by 1,116
  4. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 738
  5. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  6. 2019FreeLB: Enhanced Adversarial Training for Natural Language Understandingcited by 261
  7. 2023On the Reliability of Watermarks for Large Language Modelscited by 170
  8. 2023Platypus: Quick, Cheap, and Powerful Refinement of LLMscited by 150
  9. 2022Text-Only Training for Image Captioning using Noise-Injected CLIPcited by 91
  10. 2020Robust Optimization as Data Augmentation for Large-scale Graphscited by 58
  11. 2022ZeroPrompt: Scaling Prompt-Based Pretraining to 1,000 Tasks Improves Zero-Shot Generalizationcited by 54

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 53; 2025: 27; 2026: 19.

99 papers.

Sort:

By area:

  1. 2024Yi: Open Foundation Models by 01.AIcited by 706
  2. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertaintycited by 564
  3. 2024What matters when building vision-language models?cited by 321
  4. 2024Building and better understanding vision-language models: insights and future directionscited by 169
  5. 2024WARM: On the Benefits of Weight Averaged Reward Modelscited by 145
  6. 2024Hydra: Sequentially-Dependent Draft Heads for Medusa Decodingcited by 113
  7. 2024Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuningcited by 85
  8. 2024From LLMs to LLM-based Agents for Software Engineering: A Survey of Current, Challenges and Futurecited by 85
  9. 2024A Closer Look at the Limitations of Instruction Tuningcited by 76
  10. 2024Preserving Diversity in Supervised Fine-Tuning of Large Language Modelscited by 68
  11. 2024Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMscited by 50
  12. 2024Instruction Tuning With Loss Over Instructionscited by 40
38 more
  1. 2024IsoBench: Benchmarking Multimodal Foundation Models on Isomorphic Representationscited by 37
  2. 2024Hermes 3 Technical Reportcited by 33
  3. 2025Interpretable Steering of Large Language Models with Feature Guided Activation Additionscited by 32
  4. 2024Automating Research Synthesis with Domain-Specific Large Language Model Fine-Tuningcited by 31
  5. 2024Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Modelscited by 28
  6. 2025Roll the dice & look before you leap: Going beyond the creative limits of next-token predictioncited by 28
  7. 2024LLM-Assisted Crisis Management: Building Advanced LLM Platforms for Effective Emergency Response and Public Collaborationcited by 20
  8. 2024Masked Thought: Simply Masking Partial Reasoning Steps Can Improve Mathematical Reasoning Learning of Language Modelscited by 20
  9. 2024Aloe: A Family of Fine-tuned Open Healthcare LLMscited by 20
  10. 2024Teuken-7B-Base & Teuken-7B-Instruct: Towards European LLMscited by 20
  11. 2025SweRank: Software Issue Localization with Code Rankingcited by 20
  12. 2024LM4OPT: Unveiling the Potential of Large Language Models in Formulating Mathematical Optimization Problemscited by 18
  13. 2024A Survey on Trustworthiness in Foundation Models for Medical Image Analysiscited by 16
  14. 2024GRIN: GRadient-INformed MoEcited by 16
  15. 2024TeleChat Technical Reportcited by 15
  16. 2025Differentially Private Steering for Large Language Model Alignmentcited by 15
  17. 2025Disambiguate First, Parse Later: Generating Interpretations for Ambiguity Resolution in Semantic Parsingcited by 12
  18. 2025T-SHIRT: Token-Selective Hierarchical Data Selection for Instruction Tuningcited by 12
  19. 2024KnowTuning: Knowledge-aware Fine-tuning for Large Language Modelscited by 11
  20. 2025Diversity-driven Data Selection for Language Model Tuning through Sparse Autoencodercited by 11
  21. 2024Enhanced Automated Code Vulnerability Repair using Large Language Modelscited by 10
  22. 2024LLM Honeypot: Leveraging Large Language Models as Advanced Interactive Honeypot Systemscited by 10
  23. 2024Slight Corruption in Pre-training Data Makes Better Diffusion Modelscited by 9
  24. 2024Exploring Text-to-Motion Generation with Human Preferencecited by 8
  25. 2024Codec-ASR: Training Performant Automatic Speech Recognition Systems with Discrete Speech Representationscited by 8
  26. 2025Preference Learning with Lie Detectors can Induce Honesty or Evasioncited by 8
  27. 2024Baichuan2-Sum: Instruction Finetune Baichuan2-7B Model for Dialogue Summarizationcited by 7
  28. 2024Noise-powered Multi-modal Knowledge Graph Representation Frameworkcited by 7
  29. 2024Bridging the Training-Inference Gap in LLMs by Leveraging Self-Generated Tokenscited by 7
  30. 2024Copyright-Protected Language Generation via Adaptive Model Fusioncited by 6
  31. 2024See the Unseen: Better Context-Consistent Knowledge-Editing by Noisescited by 5
  32. 2024UniOQA: A Unified Framework for Knowledge Graph Question Answering with Large Language Modelscited by 5
  33. 2024R-SFLLM: Jamming Resilient Framework for Split Federated Learning with Large Language Modelscited by 5
  34. 2024Small Molecule Optimization with Large Language Modelscited by 5
  35. 2024mhGPT: A Lightweight Generative Pre-Trained Transformer for Mental Health Text Analysiscited by 5
  36. 2025Pensez: Less Data, Better Reasoning -- Rethinking French LLMcited by 5
  37. 2026Weak-Driven Learning: How Weak Agents make Strong Agents Strongercited by 5
  38. 2026On Surprising Effectiveness of Masking Updates in Adaptive Optimizerscited by 5

50 most cited of 99 All 99 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Shadow-FT: Tuning Instruct Model via Training on Paired Base Model
  2. 2023Instruction Tuning with GPT-4
  3. 2024Non-instructional Fine-tuning: Enabling Instruction-Following Capabilities in Pre-trained Language Models without Instruction-Following Data
  4. 2024LESS: Selecting Influential Data for Targeted Instruction Tuning
  5. 2023An Emulator for Fine-Tuning Large Language Models using Small Language Models
  6. 2023Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2
  7. 2024ReFT: Representation Finetuning for Language Models
  8. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor