Large Language Model

Large language models (LLMs) are sophisticated AI systems designed to process and generate human-like text, aiming to improve various natural language processing tasks. Current research focuses on enhancing LLM safety, efficiency (through techniques like quantization and optimized decoding), and fairness, as well as improving their ability to perform complex reasoning and handle diverse instructions. These advancements are significant because they address critical limitations in current LLMs and pave the way for broader applications across diverse fields, including healthcare, legal tech, and autonomous systems.

Papers

December 10, 2024

Identifying and Manipulating Personality Traits in LLMs Through Activation Engineering
Rumi A. Allbert, James K. Wiles, Vlad Grankovsky
Large Language Model Contrastive Learning LLM Fine Tuning LLM Explanation Personality Generation Advanced Activation Mechanism
How to Choose a Threshold for an Evaluation Metric for Large Language Models
Bhaskarjit Sarmah, Mingshu Li, Jingrao Lyu, Sebastian Frank, Nathalia Castellanos, Stefano Pasquali, Dhagash Mehta
Large Language Model Full Model Evaluation Metric Large Language Risk Assessment Canonical Threshold Faithfulness Metric LLM Based Metric
Asking Again and Again: Exploring LLM Robustness to Repeated Questions
Sagi Shaier
Large Language Model Yes No Question Complex Prompt Answer Correctness LLM Robustness Self Repetition Comprehension Datasets
Low-Rank Correction for Quantized LLMs
Meyer Scetbon, James Hensman
Large Language Model Model Compression Low Rank Language Correction Quantization Error Weight Matrix
FlexLLM: Exploring LLM Customization for Moving Target Defense on Black-Box LLMs Against Jailbreak Attacks
Bocheng Chen, Hanqing Guo, Qiben Yan
Large Language Model Jailbreak Attack Defense Method Black Box Large Language Model LLM Personalization Moving Target Defense
DRUM: Learning Demonstration Retriever for Large MUlti-modal Models
Ellen Yi-Ge, Jiechao Gao, Wei Han, Wei Zhu
Large Language Model Context Learning Large Vision Language Model Multi Modal Model Visual Language Task Demonstration Retrieval
Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs
Xiaqiang Tang, Jian Li, Nan Du, Sihong Xie
Large Language Model Knowledge Graph Retrieval Augmented Generation Multi Armed Bandit NLP Task Non Stationary Environment Retrieval Method KBQA Datasets
LLM-as-an-Interviewer: Beyond Static Testing Through Dynamic LLM Evaluation
Eunsu Kim, Juyoung Suk, Seungone Kim, Niklas Muennighoff, Dongkwan Kim, Alice Oh
Large Language Model Global Evaluation Novel Evaluation Simplicity Bias Unit Test
Hyperband-based Bayesian Optimization for Black-box Prompt Selection
Lennart Schneider, Martin Wistuba, Aaron Klein, Jacek Golebiowski, Giovanni Zappella, Felice Antonio Merra
Large Language Model Gaussian Process Bayesian Optimization Shot Example Prompt Selection
CoPrUS: Consistency Preserving Utterance Synthesis towards more realistic benchmark dialogues
Sebastian Steindl, Ulrich Schäfer, Bernd Ludwig
Large Language Model Dialogue System Dialogue Benchmark
Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM
Shaoqing Zhang, Zhuosheng Zhang, Kehai Chen, Rongxiang Weng, Muyun Yang, Tiejun Zhao, Min Zhang
Large Language Model Human Attention Jailbreak Attack Attack Success Rate Harmful Content Leap Forward Collective Perception
Optimizing Alignment with Less: Leveraging Data Augmentation for Personalized Evaluation
Javad Seraj, Mohammad Mahdi Mohajeri, Mohammad Javad Dousti, Majid Nili Ahmadabadi
Large Language Model Data Augmentation Automatic Evaluation Alignment Performance Mathematical Reasoning Task Preference Dataset
Generating Knowledge Graphs from Large Language Models: A Comparative Study of GPT-4, LLaMA 2, and BERT
Ahan Bhatt, Nandan Vaghela, Kush Dudhia
Large Language Model Knowledge Graph GPT 4 Ticket BERT Graph Domain LLaMa LlamaCare Knowledge Graph Generation
Na'vi or Knave: Jailbreaking Language Models via Metaphorical Avatars
Yu Yan, Sheng Sun, Junqi Tong, Min Liu, Qi Li
Large Language Model Language Model Metaphor Use Adversarial Challenge
ConceptSearch: Towards Efficient Program Search Using LLMs for Abstraction and Reasoning Corpus (ARC)
Kartik Singhal, Gautam Shroff
Large Language Model Concept Identification Cognitive Abstraction Reasoning Corpus
The Rise and Down of Babel Tower: Investigating the Evolution Process of Multilingual Code Large Language Model
Jiawei Chen, Wentao Chen, Jing Su, Jingjing Xu, Hongyu Lin, Mengjie Ren, Yaojie Lu, Xianpei Han, Le Sun
Large Language Model Language Specific New Language Evolutionary Process
Enhancing Relation Extraction via Supervised Rationale Verification and Feedback
Yongqi Li, Xin Miao, Shen Zhou, Mayi Xu, Yuyang Ren, Tieyun Qian
Large Language Model Human Feedback Relation Extraction Contrastive Training Feedback System
HARP: Hesitation-Aware Reframing in Transformer Inference Pass
Romain Storaï, Seung-won Hwang
Large Language Model Transformer Based Transformer Based Language Model Transformer Inference
MemHunter: Automated and Verifiable Memorization Detection at Dataset-scale in LLMs
Zhenpeng Wu, Jian Lou, Zibin Zheng, Chuan Chen
Large Language Model Training Data Task Specific Memorization Measure
KULTURE Bench: A Benchmark for Assessing Language Model in Korean Cultural Context
Xiaonan Wang, Jinyoung Yeo, Joon-Ho Lim, Hansaem Kim
Large Language Model Language Model New Benchmark Parallel Corpus Coherent Text Cultural Background Korean Ancient Document

Large Language Model

Papers

Identifying and Manipulating Personality Traits in LLMs Through Activation Engineering

How to Choose a Threshold for an Evaluation Metric for Large Language Models

Asking Again and Again: Exploring LLM Robustness to Repeated Questions

Low-Rank Correction for Quantized LLMs

FlexLLM: Exploring LLM Customization for Moving Target Defense on Black-Box LLMs Against Jailbreak Attacks

DRUM: Learning Demonstration Retriever for Large MUlti-modal Models

Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs

LLM-as-an-Interviewer: Beyond Static Testing Through Dynamic LLM Evaluation

Hyperband-based Bayesian Optimization for Black-box Prompt Selection

CoPrUS: Consistency Preserving Utterance Synthesis towards more realistic benchmark dialogues

Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM

Optimizing Alignment with Less: Leveraging Data Augmentation for Personalized Evaluation

Generating Knowledge Graphs from Large Language Models: A Comparative Study of GPT-4, LLaMA 2, and BERT

Na'vi or Knave: Jailbreaking Language Models via Metaphorical Avatars

ConceptSearch: Towards Efficient Program Search Using LLMs for Abstraction and Reasoning Corpus (ARC)

The Rise and Down of Babel Tower: Investigating the Evolution Process of Multilingual Code Large Language Model

Enhancing Relation Extraction via Supervised Rationale Verification and Feedback

HARP: Hesitation-Aware Reframing in Transformer Inference Pass

MemHunter: Automated and Verifiable Memorization Detection at Dataset-scale in LLMs

KULTURE Bench: A Benchmark for Assessing Language Model in Korean Cultural Context