research-papers

SPROUT: Authoring Programming Tutorials with Interactive Visualization of Large Language Model Generation Process

Original Paper: https://arxiv.org/abs/2312.01801 By: Yihan Liu, Zhen Wen, Luoxuan Weng, Ollie Woodman, Yi Yang, Wei Chen Abstract: The rapid development of large language models (LLMs), such as ChatGPT, has revolutionized the efficiency of creating programming tutorials. LLMs can be instructed with text prompts to generate

research-papers

Prompting AI Art: An Investigation into the Creative Skill of Prompt Engineering

Original Paper: https://arxiv.org/abs/2303.13534 By: Jonas Oppenlaender, Rhema Linder, Johanna Silvennoinen Abstract: We are witnessing a novel era of creativity where anyone can create digital content via prompt-based learning (known as prompt engineering). This paper delves into prompt engineering as a novel creative skill for creating

research-papers

Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision

Original Paper: https://arxiv.org/abs/2305.03047 By: Zhiqing Sun, Yikang Shen, Qinhong Zhou, Hongxin Zhang, Zhenfang Chen, David Cox, Yiming Yang, Chuang Gan Abstract: Recent AI-assistant agents, such as ChatGPT, predominantly rely on supervised fine-tuning (SFT) with human annotations and reinforcement learning from human feedback (RLHF) to align

research-papers

ImageDream: Image-Prompt Multi-view Diffusion for 3D Generation

Original Paper: https://arxiv.org/abs/2312.02201 By: Peng Wang, Yichun Shi Abstract: We introduce "ImageDream," an innovative image-prompt, multi-view diffusion model for 3D object generation. ImageDream stands out for its ability to produce 3D models of higher quality compared to existing state-of-the-art, image-conditioned methods. Our approach

research-papers

What we learned from speaking to 50+ LLM developers building RAG apps.

We’ve spoken to over 50 developers building RAG-based applications last month. Here are 5 Common Mistakes an LLM might make: 1. Bad Retrievals: This normally happens when the context provided is irrelevant to the query. 2. Unfaithful Responses: Often, the response doesn't stay true to the context

research-papers

PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification

Original Paper: https://arxiv.org/abs/2308.02816 By: Hongwei Yao, Jian Lou, Kui Ren, Zhan Qin Abstract: Large language models (LLMs) have witnessed a meteoric rise in popularity among the general public users over the past few months, facilitating diverse downstream tasks with human-level accuracy and proficiency. Prompts play

research-papers

Deficiency of Large Language Models in Finance: An Empirical Examination of Hallucination

Original Paper: https://arxiv.org/abs/2311.15548 By: Haoqiang Kang, Xiao-Yang Liu Abstract: The hallucination issue is recognized as a fundamental deficiency of large language models (LLMs), especially when applied to fields such as finance, education, and law. Despite the growing concerns, there has been a lack of empirical

research-papers

How to Prompt LLMs for Text-to-SQL: A Study in Zero-shot, Single-domain, and Cross-domain Settings

Original Paper: https://arxiv.org/abs/2305.11853 By: Shuaichen Chang, Eric Fosler-Lussier Abstract: Large language models (LLMs) with in-context learning have demonstrated remarkable capability in the text-to-SQL task. Previous research has prompted LLMs with various demonstration-retrieval strategies and intermediate reasoning steps to enhance the performance of LLMs. However, those

research-papers

Probabilistic Tree-of-thought Reasoning for Answering Knowledge-intensive Complex Questions

Original Paper: https://arxiv.org/abs/2311.13982 By: Shulin Cao, Jiajie Zhang, Jiaxin Shi, Xin Lv, Zijun Yao, Qi Tian, Juanzi Li, Lei Hou Abstract: Large language models (LLMs) are capable of answering knowledge-intensive complex questions with chain-of-thought (CoT) reasoning. However, they tend to generate factually incorrect reasoning steps

research-papers

Prompt a Robot to Walk with Large Language Models

Original Paper: https://arxiv.org/abs/2309.09969 By: Yen-Jen Wang, Bike Zhang, Jianyu Chen, Koushil Sreenath Abstract: Large language models (LLMs) pre-trained on vast internet-scale data have showcased remarkable capabilities across diverse domains. Recently, there has been escalating interest in deploying LLMs for robotics, aiming to harness the power

research-papers

To be or not to be? an exploration of continuously controllable prompt engineering

Original Paper: https://arxiv.org/abs/2311.09773 By: Yuhan Sun, Mukai Li, Yixin Cao, Kun Wang, Wenxiao Wang, Xingyu Zeng, Rui Zhao Abstract: As the use of large language models becomes more widespread, techniques like parameter-efficient fine-tuning and other methods for controlled generation are gaining traction for customizing models

research-papers

Fine-tuning Language Models for Factuality

Original Paper: https://arxiv.org/abs/2311.08401 By: Katherine Tian, Eric Mitchell, Huaxiu Yao, Christopher D. Manning, Chelsea Finn Abstract: The fluency and creativity of large pre-trained language models (LLMs) have led to their widespread use, sometimes even as a replacement for traditional search engines. Yet language models are