Athina AI

research-papers

Prompt Injection attack against LLM-integrated Applications

Original Paper: https://arxiv.org/abs/2306.05499 By: Yi Liu, Gelei Deng, Yuekang Li, Kailong Wang, Zihao Wang, Xiaofeng Wang, Tianwei Zhang, Yepang Liu, Haoyu Wang, Yan Zheng, Yang Liu Abstract: Large Language Models (LLMs), renowned for their superior proficiency in language comprehension and generation, stimulate a vibrant ecosystem

research-papers

Inferring Properties of Graph Neural Networks

Original Paper: https://arxiv.org/abs/2401.03790 By: Dat Nguyen (1), Hieu M. Vu (2), Cong-Thanh Le (1), Bach Le (1), David Lo (3), ThanhVu Nguyen (4)Corina Pasareanu (5) ((1) University of Melbourne, (2) Independent Researcher, (3) Singapore Management University, (4) George Mason University, (5) Carnegie Mellon University)

research-papers

Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

Original Paper: https://arxiv.org/abs/2310.02304 By: Eric Zelikman, Eliana Lorch, Lester Mackey, Adam Tauman Kalai Abstract: Several recent advances in AI systems (e.g., Tree-of-Thoughts and Program-Aided Language Models) solve problems by providing a "scaffolding" program that structures multiple calls to language models to generate

research-papers

Consistency-guided Prompt Learning for Vision-Language Models

Original Paper: https://arxiv.org/abs/2306.01195 By: Shuvendu Roy, Ali Etemad Abstract: We propose Consistency-guided Prompt learning (CoPrompt), a new fine-tuning method for vision-language models. Our approach improves the generalization of large foundation models when fine-tuned on downstream tasks in a few-shot setting. The basic idea of CoPrompt

research-papers

Connecting Large Language Models with Evolutionary Algorithms Yields Powerful Prompt Optimizers

Original Paper: https://arxiv.org/abs/2309.08532 By: Qingyan Guo, Rui Wang, Junliang Guo, Bei Li, Kaitao Song, Xu Tan, Guoqing Liu, Jiang Bian, Yujiu Yang Abstract: Large Language Models (LLMs) excel in various tasks, but they rely on carefully crafted prompts that often demand substantial human effort. To

research-papers

Everything of Thoughts: Defying the Law of Penrose Triangle for Thought Generation

Original Paper: https://arxiv.org/abs/2311.04254 By: Ruomeng Ding, Chaoyun Zhang, Lu Wang, Yong Xu, Minghua Ma, Wei Zhang, Si Qin, Saravan Rajmohan, Qingwei Lin, Dongmei Zhang Abstract: Recent advancements in Large Language Models (LLMs) have revolutionized decision-making by breaking down complex problems into more manageable language sequences

research-papers

Tree of Attacks: Jailbreaking Black-Box LLMs Automatically

Original Paper: https://arxiv.org/abs/2312.02119 By: Anay Mehrotra, Manolis Zampetakis, Paul Kassianik, Blaine Nelson, Hyrum Anderson, Yaron Singer, Amin Karbasi Abstract: While Large Language Models (LLMs) display versatile functionality, they continue to generate harmful, biased, and toxic content, as demonstrated by the prevalence of human-designed jailbreaks. In

research-papers

Gemma: Open Models Based on Gemini, Research and Technology

Original Paper: https://storage.googleapis.com/deepmind-media/gemma/gemma-report.pdf By: Gemma Team, Google DeepMind1 Abstract: This work introduces Gemma, a family of lightweight, state-of-the art open models built from the research and technology used to create Gemini models. Gemma models demonstrate strong performance across academic benchmarks for language understanding,

research-papers

Prompt Engineering a Prompt Engineer

Original Paper: https://arxiv.org/abs/2311.05661 By: Qinyuan Ye, Maxamed Axmed, Reid Pryzant, Fereshte Khani Abstract: Prompt engineering is a challenging yet crucial task for optimizing the performance of large language models on customized tasks. It requires complex reasoning to examine the model's errors, hypothesize what

research-papers

GTBench: Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluations

Original Paper: https://arxiv.org/abs/2402.12348 By: Jinhao Duan, Renming Zhang, James Diffenderfer, Bhavya Kailkhura, Lichao Sun, Elias Stengel-Eskin, Mohit Bansal, Tianlong Chen, Kaidi Xu Abstract: As Large Language Models (LLMs) are integrated into critical real-world applications, their strategic and logical reasoning abilities are increasingly crucial. This paper

research-papers

Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information

Original Paper: https://arxiv.org/abs/2311.11509 By: Zhengmian Hu, Gang Wu, Saayan Mitra, Ruiyi Zhang, Tong Sun, Heng Huang, Viswanathan Swaminathan Abstract: In recent years, Large Language Models (LLM) have emerged as pivotal tools in various applications. However, these models are susceptible to adversarial prompt attacks, where attackers

research-papers

DePT: Decomposed Prompt Tuning for Parameter-Efficient Fine-tuning

Original Paper: https://arxiv.org/abs/2309.05173 By: Zhengxiang Shi, Aldo Lipani Abstract: Prompt tuning (PT), where a small amount of trainable soft (continuous) prompt vectors is affixed to the input of language models (LM), has shown promising results across various tasks and models for parameter-efficient fine-tuning (PEFT). PT