paper-with-me

Papers

Trusted Source Alignment in Large Language Models

2023-11-12 · Vasilisa Bashlovkina, Zhaobin Kuang, Riley Matthews, Edward Clifford, Yennie Jun, William W. Cohen, Simon Baumgartner

Large language models (LLMs) are trained on web-scale corpora that inevitably include contradictory factual information from sources of varying reliability. In this paper, we propose measuring an LLM property called trusted source alignment (TSA): the model's propensity to align with content produced by trusted publishers in the face of uncertainty or controversy. We present FactCheckQA, a TSA evaluation dataset based on a corpus of fact checking articles. We describe a simple protocol for evaluating TSA and offer a detailed analysis of design considerations including response extraction, claim contextualization, and bias in prompt formulation. Applying the protocol to PaLM-2, we find that as we scale up the model size, the model performance on FactCheckQA improves from near-random to up to 80% balanced accuracy in aligning with trusted sources.

📄 PDF Abstract BibTeX arXiv:2311.06697

Code (0)

등록된 구현이 없습니다.

Tasks

ArticlesFact Checking

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats

2024-11-26 · Jiaxin Wen, Vivek Hebbar, Caleb Larson, Aryan Bhatt 외

As large language models (LLMs) become increasingly capable, it is prudent to assess whether safety measures remain effective even if LLMs intentionally try to bypass them. Previous work introduced control evaluations, a…

Code Generation

TraceGuard: Structured Multi-Dimensional Monitoring as a Collusion-Resistant Control Protocol

2026-04-05 · Khanh Linh Nguyen, Hoa Nghiem, Tu Tran arxiv

AI control protocols use monitors to detect attacks by untrusted AI agents, but standard single-score monitors face two limitations: they miss subtle attacks where outputs look clean but reasoning is off, and they collap…

Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noise

2018-02-14 · NeurIPS 2018 12 · Dan Hendrycks, Mantas Mazeika, Duncan Wilson, Kevin Gimpel

The growing importance of massive datasets used for deep learning makes robustness to label noise a critical property for classifiers to have. Sources of label noise include automatic labeling, non-expert labeling, and l…

Data Poisoning

Detecting fake news for the new coronavirus by reasoning on the Covid-19 ontology

2020-04-26 · Adrian Groza

In the context of the Covid-19 pandemic, many were quick to spread deceptive information. I investigate here how reasoning in Description Logics (DLs) can detect inconsistencies between trusted medical sources and not tr…

Cross-Region Domain Adaptation for Class-level Alignment

2021-09-14 · Zhijie Wang, Xing Liu, Masanori Suganuma, Takayuki Okatani

Semantic segmentation requires a lot of training data, which necessitates costly annotation. There have been many studies on unsupervised domain adaptation (UDA) from one domain to another, e.g., from computer graphics t…

Domain AdaptationImage-to-Image TranslationSemantic SegmentationSynthetic-to-Real Translation+1