# Latent Fact-Checking: Detecting Misinformation through Activation Engineering

arXiv: https://arxiv.org/abs/2608.06417 · HTML: https://mlsanity.com/paper/2608.06417/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 27 references that are in the map.
- First posted on arXiv 5 Aug 2026.

## Builds on (most cited first, 23 of 23)

- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [FEVER: a large-scale dataset for Fact Extraction and VERification](https://mlsanity.com/paper/1803.05355/) (2018)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [The Linear Representation Hypothesis and the Geometry of Large Language Models](https://mlsanity.com/paper/2311.03658/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [Fact or Fiction: Verifying Scientific Claims](https://mlsanity.com/paper/2004.14974/) (2020)
- ["Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection](https://mlsanity.com/paper/1705.00648/) (2017)
- [Extracting Latent Steering Vectors from Pretrained Language Models](https://mlsanity.com/paper/2205.05124/) (2022)
- [FEVEROUS: Fact Extraction and VERification Over Unstructured and Structured information](https://mlsanity.com/paper/2106.05707/) (2021)
- [AVeriTeC: A Dataset for Real-world Claim Verification with Evidence from the Web](https://mlsanity.com/paper/2305.13117/) (2023)
- [Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization](https://mlsanity.com/paper/2406.00045/) (2024)
- [Style Vectors for Steering Generative Large Language Model](https://mlsanity.com/paper/2402.01618/) (2024)
- [Controlling Large Language Models Through Concept Activation Vectors](https://mlsanity.com/paper/2501.05764/) (2025)
- [Multimodal Misinformation Detection by Learning from Synthetic Data with Multimodal LLMs](https://mlsanity.com/paper/2409.19656/) (2024)
- [Analysis of Disinformation and Fake News Detection Using Fine-Tuned Large Language Model](https://mlsanity.com/paper/2309.04704/) (2023)
- [AIC CTU system at AVeriTeC: Re-framing automated fact-checking as a simple RAG task](https://mlsanity.com/paper/2410.11446/) (2024)
- [FACTors: A New Dataset for Studying the Fact-checking Ecosystem](https://mlsanity.com/paper/2505.09414/) (2025)
- [Zero-Shot Learning and Key Points Are All You Need for Automated Fact-Checking](https://mlsanity.com/paper/2408.08400/) (2024)

## Abstract (arXiv)

The proliferation of misinformation online has driven demand for scalable detection systems. While most existing approaches rely on surface-level linguistic features or external knowledge retrieval, we examine truthfulness as a geometric property of a language model's representation space. We introduce a misinformation detection framework grounded in activation engineering, which leverages the latent geometry of transformer models. Our approach elicits a misinformation direction in the residual stream by contrasting activations from paired truthful and false statements, following the difference-in-means principle of Contrastive Activation Addition (CAA). At inference time, the last-token activation of an unseen claim is projected onto this direction, and the projected representation is fed to an Multilayer Perceptron (MLP) for classification. The procedure requires no fine-tuning of the backbone model, no external evidence retrieval, and no task-specific supervision beyond the contrastive pairs used to estimate the direction. We evaluate the method across 11 models from the Gemma, Llama, and Qwen families, ranging from 270M to 12B parameters, on three fact-checking benchmarks: AVeriTeC, LIAR, and FACTors. The falsehood direction is recoverable across model scales and architectural families, and last-token projection matches or surpasses zero-shot and few-shot prompting baselines on LIAR and FACTors, with the largest gains observed for smaller models. Performance on AVeriTeC is more limited, which we attribute to its evidence-grounded labeling scheme. These findings provide evidence that truthfulness is a structured, linearly separable concept in the latent space of pretrained language models, and point toward interpretability-driven misinformation detection as a practical complement to retrieval-based pipelines. The code is available on https://github.com/Malta-Lab/LaFaCt.