A survey of large language models

WX Zhao, K Zhou, J Li, T Tang, X Wang, Y Hou… - arXiv preprint arXiv …, 2023 - arxiv.org
Language is essentially a complex, intricate system of human expressions governed by
grammatical rules. It poses a significant challenge to develop capable AI algorithms for …

Unleashing the emergent cognitive synergy in large language models: A task-solving agent through multi-persona self-collaboration

Z Wang, S Mao, W Wu, T Ge, F Wei, H Ji - arXiv preprint arXiv:2307.05300, 2023 - arxiv.org
Human intelligence thrives on the concept of cognitive synergy, where collaboration and
information integration among different cognitive processes yield superior outcomes …

Self-verification improves few-shot clinical information extraction

Z Gero, C Singh, H Cheng, T Naumann… - arXiv preprint arXiv …, 2023 - arxiv.org
Extracting patient information from unstructured text is a critical task in health decision-
support and clinical research. Large language models (LLMs) have shown the potential to …

Enable language models to implicitly learn self-improvement from data

Z Wang, L Hou, T Lu, Y Wu, Y Li, H Yu, H Ji - arXiv preprint arXiv …, 2023 - arxiv.org
Large Language Models (LLMs) have demonstrated remarkable capabilities in open-ended
text generation tasks. However, the inherent open-ended nature of these tasks implies that …

Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?

Q Wang, Z Wang, Y Su, H Tong, Y Song - arXiv preprint arXiv:2402.18272, 2024 - arxiv.org
Recent progress in LLMs discussion suggests that multi-agent discussion improves the
reasoning abilities of LLMs. In this work, we reevaluate this claim through systematic …

Literature Review of AI Hallucination Research Since the Advent of ChatGPT: Focusing on Papers from arXiv

DM Park, HJ Lee - Informatization Policy, 2024 - koreascience.kr
Hallucination is a significant barrier to the utilization of large-scale language models or
multimodal models. In this study, we collected 654 computer science papers with" …

Autohall: Automated hallucination dataset generation for large language models

Z Cao, Y Yang, H Zhao - arXiv preprint arXiv:2310.00259, 2023 - arxiv.org
While Large language models (LLMs) have garnered widespread applications across
various domains due to their powerful language understanding and generation capabilities …

Natural language deduction with incomplete information

Z Sprague, K Bostrom, S Chaudhuri… - arXiv preprint arXiv …, 2022 - arxiv.org
A growing body of work studies how to answer a question or verify a claim by generating a
natural language" proof": a chain of deductive inferences yielding the answer based on a set …

Parameter-efficient tuning helps language model alignment

T Xue, Z Wang, H Ji - arXiv preprint arXiv:2310.00819, 2023 - arxiv.org
Aligning large language models (LLMs) with human preferences is essential for safe and
useful LLMs. Previous works mainly adopt reinforcement learning (RLHF) and direct …

Forward-backward reasoning in large language models for mathematical verification

W Jiang, H Shi, L Yu, Z Liu, Y Zhang, Z Li… - Findings of the …, 2024 - aclanthology.org
Self-Consistency samples diverse reasoning chains with answers and chooses the final
answer by majority voting. It is based on forward reasoning and cannot further improve …