REVIEW 4 cited by
Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
The knowledge within large language models (LLMs) may become outdated quickly. While in-context editing (ICE) is currently the most effective method for knowledge editing (KE), it is constrained by the black-box modeling of LLMs and thus lacks interpretability. Our work aims to elucidate the superior performance of ICE on the KE by analyzing the impacts of in-context new knowledge on token-wise distributions. We observe that despite a significant boost in logits of the new knowledge, the performance of is still hindered by stubborn knowledge. Stubborn knowledge refers to as facts that have gained excessive confidence during pretraining, making it hard to edit effectively. To address this issue and further enhance the performance of ICE, we propose a novel approach termed $\textbf{De}$coding by $\textbf{C}$ontrasting $\textbf{K}$nowledge (DeCK). DeCK derives the distribution of the next token by contrasting the logits obtained from the newly edited knowledge guided by ICE with those from the unedited parametric knowledge. Our experiments consistently demonstrate that DeCK enhances the confidence of LLMs in edited facts. For instance, it improves the performance of LLaMA3-8B-instruct on MQuAKE by up to 219%, demonstrating its capability to strengthen ICE in the editing of stubborn knowledge. Our work paves the way to develop the both effective and accountable KE methods for LLMs. (The source code is available at: https://deck-llm.meirtz.com)
Forward citations
Cited by 4 Pith papers
-
DICA: Dual-Indicator Guided Contrastive Alignment in Multimodal Large Language Models
Monitoring attention entropy and image-output correlation during decoding, then applying targeted contrastive corrections, reduces hallucination in multimodal LLMs without retraining.
-
A Survey on Proactive Defense Strategies Against Misinformation in Large Language Models
A survey claims proactive defenses against LLM misinformation outperform post-hoc detection by up to 63%, but no meta-analysis details are provided to support the claim.
-
Large Language Models as Computable Approximations to Solomonoff Induction
The paper argues LLMs are computable approximations of Solomonoff induction, but its central derivation recovers the model's own probabilities by construction.
-
Tutorial on Using Machine Learning and Deep Learning Models for Mental Illness Detection
A tutorial-style benchmark showing that standard classifiers reach binary F1 0.93 to 0.96 and multiclass F1 0.75 to 0.78 on a Kaggle mental-health text dataset, with no new method or result introduced.
Discussion (0). Continue with ORCID to comment.