Knowledge Conflicts for LLMs: A Survey
- URL: http://arxiv.org/abs/2403.08319v2
- Date: Sat, 22 Jun 2024 08:31:40 GMT
- Title: Knowledge Conflicts for LLMs: A Survey
- Authors: Rongwu Xu, Zehan Qi, Zhijiang Guo, Cunxiang Wang, Hongru Wang, Yue Zhang, Wei Xu,
- Abstract summary: Survey focuses on three categories of knowledge conflicts: context-memory, inter-context, and intra-memory conflict.
These conflicts can significantly impact the trustworthiness and performance of large language models.
- Score: 24.731074825915833
- License: http://creativecommons.org/licenses/by-nc-nd/4.0/
- Abstract: This survey provides an in-depth analysis of knowledge conflicts for large language models (LLMs), highlighting the complex challenges they encounter when blending contextual and parametric knowledge. Our focus is on three categories of knowledge conflicts: context-memory, inter-context, and intra-memory conflict. These conflicts can significantly impact the trustworthiness and performance of LLMs, especially in real-world applications where noise and misinformation are common. By categorizing these conflicts, exploring the causes, examining the behaviors of LLMs under such conflicts, and reviewing available solutions, this survey aims to shed light on strategies for improving the robustness of LLMs, thereby serving as a valuable resource for advancing research in this evolving area.
Related papers
- From Internal Conflict to Contextual Adaptation of Language Models [42.776896363518844]
Studies indicate that LMs often ignore the provided context as it can conflict with the pre-existing LM's memory learned during pre-training.
We introduce the DYNAMICQA dataset, which includes facts with a temporal dynamic nature.
Our experiments reveal that static facts, which are unlikely to change, are more easily updated with additional context.
arXiv Detail & Related papers (2024-07-24T06:06:07Z) - A Survey on Large Language Models for Critical Societal Domains: Finance, Healthcare, and Law [65.87885628115946]
Large language models (LLMs) are revolutionizing the landscapes of finance, healthcare, and law.
We highlight the instrumental role of LLMs in enhancing diagnostic and treatment methodologies in healthcare, innovating financial analytics, and refining legal interpretation and compliance strategies.
We critically examine the ethics for LLM applications in these fields, pointing out the existing ethical concerns and the need for transparent, fair, and robust AI systems.
arXiv Detail & Related papers (2024-05-02T22:43:02Z) - CausalBench: A Comprehensive Benchmark for Causal Learning Capability of Large Language Models [27.362012903540492]
Causality reveals fundamental principles behind data distributions in real-world scenarios.
Large language models (LLMs) can understand causality directly impacts their efficacy across explaining outputs, adapting to new evidence, and generating counterfactuals.
This paper proposes a comprehensive benchmark, namely CausalBench, to evaluate the causality understanding capabilities of LLMs.
arXiv Detail & Related papers (2024-04-09T14:40:08Z) - Untangle the KNOT: Interweaving Conflicting Knowledge and Reasoning Skills in Large Language Models [51.72963030032491]
Knowledge documents for large language models (LLMs) may conflict with the memory of LLMs due to outdated or incorrect knowledge.
We construct a new dataset, dubbed KNOT, for knowledge conflict resolution examination in the form of question answering.
arXiv Detail & Related papers (2024-04-04T16:40:11Z) - LLM as a Mastermind: A Survey of Strategic Reasoning with Large Language Models [75.89014602596673]
Strategic reasoning requires understanding and predicting adversary actions in multi-agent settings while adjusting strategies accordingly.
We explore the scopes, applications, methodologies, and evaluation metrics related to strategic reasoning with Large Language Models.
It underscores the importance of strategic reasoning as a critical cognitive capability and offers insights into future research directions and potential improvements.
arXiv Detail & Related papers (2024-04-01T16:50:54Z) - Large Language Models and Causal Inference in Collaboration: A Comprehensive Survey [46.4375135354838]
Causal inference has shown potential in enhancing the predictive accuracy, fairness, robustness, and explainability of Natural Language Processing (NLP) models.
The emergence of generative Large Language Models (LLMs) has significantly impacted various NLP domains.
arXiv Detail & Related papers (2024-03-14T17:47:20Z) - Tug-of-War Between Knowledge: Exploring and Resolving Knowledge
Conflicts in Retrieval-Augmented Language Models [18.82042974470535]
Retrieval-augmented language models (RALMs) have demonstrated significant potential in refining and expanding their internal memory.
Knowledge conflicts can ensnare RALMs in a tug-of-war between knowledge, limiting their practical applicability.
We propose a method called Conflict-Disentangle Contrastive Decoding (CD2) to better calibrate the model's confidence.
arXiv Detail & Related papers (2024-02-22T09:51:08Z) - Resolving Knowledge Conflicts in Large Language Models [48.92369530853202]
Large language models (LLMs) often encounter knowledge conflicts.
We ask what are the desiderata for LLMs when a knowledge conflict arises and whether existing LLMs fulfill them.
We introduce KNOWLEDGE CONFLICT, an evaluation framework for simulating contextual knowledge conflicts.
arXiv Detail & Related papers (2023-10-02T06:57:45Z) - Investigating the Factual Knowledge Boundary of Large Language Models
with Retrieval Augmentation [91.30946119104111]
We show that large language models (LLMs) possess unwavering confidence in their capabilities to respond to questions.
Retrieval augmentation proves to be an effective approach in enhancing LLMs' awareness of knowledge boundaries.
We also find that LLMs have a propensity to rely on the provided retrieval results when formulating answers.
arXiv Detail & Related papers (2023-07-20T16:46:10Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.