LLM Based Augmentation

LLM-based augmentation leverages the generative capabilities of large language models to expand and enhance existing datasets for various natural language processing tasks. Current research focuses on comparing LLM augmentation's cost-effectiveness against established methods, optimizing data selection strategies for improved cross-lingual performance, and mitigating risks like the propagation of biases through self-referential learning loops. This approach holds significant promise for improving model accuracy in low-resource settings and augmenting human performance in tasks like forecasting, but careful consideration of potential drawbacks, such as the overreliance on generated contexts, is crucial for responsible implementation.

Papers

October 12, 2024

ELICIT: LLM Augmentation via External In-Context Capability
Futing Wang, Jianhao Yan, Yue Zhang, Tao Lin
Large Language Model Context Learning Task Vector Eliciting Knowledge Elicitation Method LLM Based Augmentation

October 11, 2024

JurEE not Judges: safeguarding llm interactions with small, specialised Encoder Ensembles
Dom Nasrabadi
Chatbot Response Encoder Model Expert Tax Judge Encoder Only Transformer LLM Based Augmentation Ensemble Encoding Model

August 29, 2024

LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?
Jan Cegin, Jakub Simko, Peter Brusilovsky
Large Language Model Classification Code LLM Based Complementary Benefit Generative Large Language Model Hidden CoST LLM Based Augmentation

July 15, 2024

Boosting Zero-Shot Crosslingual Performance using LLM-Based Augmentations with Effective Data Selection
Barah Fazili, Ashish Sunil Agrawal, Preethi Jyothi
Large Language Model Cross Lingual Transfer Data Selection Natural Language Inference Task Cross Lingual Performance LLM Based Augmentation

March 15, 2024

Large Language Models and User Trust: Consequence of Self-Referential Learning Loop and the Deskilling of Healthcare Professionals
Avishek Choudhury, Zaria Chaudhry
Generated Content User Trust LLM Generated Solution Healthcare Professional LLM Based Augmentation Clinician Friendly Solution Training Loop

February 12, 2024

AI-Augmented Predictions: LLM Assistants Improve Human Forecasting Accuracy
Philipp Schoenegger, Peter S. Park, Ezra Karger, Philip E. Tetlock
Large Language Model Human Prediction Forecasting Performance Forecasting Task LLM Powered Writing LLM Based Augmentation

January 22, 2024

Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts?
Hexiang Tan, Fei Sun, Wanli Yang, Yuanzhuo Wang, Qi Cao, Xueqi Cheng
Language Model Context Information Knowledge Conflict Context Retrieval Context Generation LLM Based Augmentation Retrieval Augmented LLM

January 15, 2024

JumpCoder: Go Beyond Autoregressive Coder via Online Modification
Mouxiang Chen, Hao Tian, Zhongxin Liu, Xiaoxue Ren, Jianling Sun
Code Generation Code LLM LLM Based Augmentation Online Change Autoregressive Coding

January 4, 2024

LLM Augmented LLMs: Expanding Capabilities through Composition
Rachit Bansal, Bidisha Samanta, Siddharth Dalmia, Nitish Gupta, Shikhar Vashishth, Sriram Ganapathy, Abhishek Bapna, Prateek Jain, Partha Talukdar
Language Model Code Generation Low Resource Language Capability Evolution Compositional Ability Foundational Model Code Model LLM Based Augmentation

September 1, 2023

FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning
Weirui Kuang, Bingchen Qian, Zitao Li, Daoyuan Chen, Dawei Gao, Xuchen Pan, Yuexiang Xie, Yaliang Li, Bolin Ding, Jingren Zhou
Fine Tuning LLM Fine Tuning Federated Setting LLM Benchmark Fine Tuned LLM SPR Package LLM Based Augmentation

July 14, 2023

Controllable Data Augmentation for Few-Shot Text Mining with Chain-of-Thought Attribute Manipulation
Letian Peng, Yuwei Zhang, Jingbo Shang
Large Language Model Data Augmentation Text Mining Attribute Manipulation LLM Based Augmentation Shot Natural Language Understanding

February 24, 2023

Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback
Baolin Peng, Michel Galley, Pengcheng He, Hao Cheng, Yujia Xie, Yu Hu, Qiuyuan Huang, Lars Liden, Zhou Yu, Weizhu Chen, Jianfeng Gao
Large Language Model Task Oriented External Knowledge Multi FAct LLM Based Augmentation

February 19, 2022

LPC Augment: An LPC-Based ASR Data Augmentation Algorithm for Low and Zero-Resource Children's Dialects
Alexander Johnson, Ruchao Fan, Robin Morris, Abeer Alwan
Automatic Speech Recognition Low Resource Automatic Speech Recognition Performance Low Priority Speech Datasets Regional Dialect DNN HMM LLM Based Augmentation

LLM Based Augmentation

Papers

ELICIT: LLM Augmentation via External In-Context Capability

JurEE not Judges: safeguarding llm interactions with small, specialised Encoder Ensembles

LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?

Boosting Zero-Shot Crosslingual Performance using LLM-Based Augmentations with Effective Data Selection

Large Language Models and User Trust: Consequence of Self-Referential Learning Loop and the Deskilling of Healthcare Professionals

AI-Augmented Predictions: LLM Assistants Improve Human Forecasting Accuracy

Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts?

JumpCoder: Go Beyond Autoregressive Coder via Online Modification

LLM Augmented LLMs: Expanding Capabilities through Composition

FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning

Controllable Data Augmentation for Few-Shot Text Mining with Chain-of-Thought Attribute Manipulation

Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback

LPC Augment: An LPC-Based ASR Data Augmentation Algorithm for Low and Zero-Resource Children's Dialects