相关论文: Locating and Editing Factual Associations in GPT

Recent advances in Knowledge Editing (KE), particularly Rank-One Model Editing (ROME), show superior efficiency over fine-tuning and in-context learning for updating single-hop facts in transformers. However, these methods face significant…

Knowledge editing aims to update outdated information in Large Language Models (LLMs). A representative line of study is locate-then-edit methods, which typically employ causal tracing to identify the modules responsible for recalling…

The locate-then-edit paradigm has shown significant promise for knowledge editing (KE) in Large Language Models (LLMs). While previous methods perform well on single-hop fact recall tasks, they consistently struggle with multi-hop factual…

Model editing techniques modify a minor proportion of knowledge in Large Language Models (LLMs) at a relatively low cost, which have demonstrated notable success. Existing methods assume Transformer Layer (TL) hidden states are values of…

Pretrained language models can encode a large amount of knowledge and utilize it for various reasoning tasks, yet they can still struggle to learn novel factual knowledge effectively from finetuning on limited textual demonstrations. In…

Large Audio-Language Models (LALMs) have shown strong performance in speech understanding, making speech a natural interface for accessing factual information. Yet they are trained on static corpora and may encode incorrect facts. Existing…

Recent work using Rank-One Model Editing (ROME), a popular model editing method, has shown that there are certain facts that the algorithm is unable to edit without breaking the model. Such edits have previously been called disabling edits.…

Truth discovery is to resolve conflicts and find the truth from multiple-source statements. Conventional methods mostly research based on the mutual effect between the reliability of sources and the credibility of statements, however, pay…

Large language models (LLMs) struggle with maintaining accurate knowledge due to conflicting/outdated parametric memories. While locate-and-edit methods address this, their reliance on models' internal representations leads to robustness…

ROME and MEMIT are largely believed to be two different model editing algorithms, with the major difference between them being the ability to perform batched edits. In this paper, we unify these two algorithms under a single conceptual…

The feed-forward networks (FFNs) in transformers are recognized as a group of key-value neural memories to restore abstract high-level knowledge. In this work, we conduct an empirical ablation study on updating keys (the 1st layer in the…

Factual incorrectness in generated content is one of the primary concerns in ubiquitous deployment of large language models (LLMs). Prior findings suggest LLMs can (sometimes) detect factual incorrectness in their generated content (i.e.,…

Abstractive summarization models often generate factually inconsistent content particularly when the parametric knowledge of the model conflicts with the knowledge in the input document. In this paper, we analyze the robustness of…

Pretrained language models sometimes possess knowledge that we do not wish them to, including memorized personal information and knowledge that could be used to harm people. They can also output toxic or harmful text. To mitigate these…

We present a novel extension to Retrieval Augmented Generation with the goal of mitigating factual inaccuracies in the output of large language models. Specifically, our method draws on the cognitive linguistic theory of frame semantics for…

Recent research suggests that the feed-forward module within Transformers can be viewed as a collection of key-value memories, where the keys learn to capture specific patterns from the input based on the training examples. The values then…

We present a reproduction study of "Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals" (Ortu et al., 2024), which investigates competition of mechanisms in language models between factual recall and…

Continual Pre-Training (CPT) is essential for enabling Language Models (LMs) to integrate new knowledge without erasing old. While classical CPT techniques like data replay have become the standard paradigm, the mechanisms underlying how…

How do transformer language models memorize factual associations? A common view casts internal weight matrices as associative memories over pairs of embeddings, requiring parameter counts that scale linearly with the number of facts. We…

Multilingual pretrained language models serve as repositories of multilingual factual knowledge. Nevertheless, a substantial performance gap of factual knowledge probing exists between high-resource languages and low-resource languages,…