# From Cheap Fakes to Pure Synthesis: Addressing the New Era of T2V Fake News Videos

arXiv: https://arxiv.org/abs/2608.06732 · HTML: https://mlsanity.com/paper/2608.06732/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 48 references that are in the map.
- First posted on arXiv 7 Aug 2026.

## Builds on (most cited first, 18 of 18)

- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [SNIFFER: Multimodal Large Language Model for Explainable Out-of-Context Misinformation Detection](https://mlsanity.com/paper/2403.03170/) (2024)
- [Video-Bench: Human-Aligned Video Generation Benchmark](https://mlsanity.com/paper/2504.04907/) (2025)
- [Explainable Fake News Detection With Large Language Model via Defense Among Competing Wisdom](https://mlsanity.com/paper/2405.03371/) (2024)
- [SurrogatePrompt: Bypassing the Safety Filter of Text-to-Image Models via Substitution](https://mlsanity.com/paper/2309.14122/) (2023)
- [FakingRecipe: Detecting Fake News on Short Video Platforms from the Perspective of Creative Process](https://mlsanity.com/paper/2407.16670/) (2024)
- [Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning](https://mlsanity.com/paper/2512.15693/) (2025)
- [Two Heads Are Better Than One: Improving Fake News Video Detection by Correlating with Neighbors](https://mlsanity.com/paper/2306.05241/) (2023)
- [Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoning](https://mlsanity.com/paper/2505.16836/) (2025)
- [Similarity over Factuality: Are we making progress on multimodal out-of-context misinformation detection?](https://mlsanity.com/paper/2407.13488/) (2024)
- [StealthGraph: Exposing Domain-Specific Risks in LLMs through Knowledge-Graph-Guided Harmful Prompt Generation](https://mlsanity.com/paper/2601.04740/) (2026)
- [Content-based Analysis of the Cultural Differences between TikTok and Douyin](https://mlsanity.com/paper/2011.01414/) (2020)
- [Enhancing Fake News Video Detection via LLM-Driven Creative Process Simulation](https://mlsanity.com/paper/2510.04024/) (2025)

## Built on by (most cited first, 1 of 1)

- [Detecting and Explaining Fake News Short Videos with Multimodal Content and Real-World Evidence](https://mlsanity.com/paper/2609.12678/) (2026)

## Abstract (arXiv)

Recent text-to-video (T2V) generation models enable fake news videos to be synthesized from scratch, shifting the threat beyond cheap fakes assembled from existing footage. Such news videos can closely match fabricated narratives, creating a modality alignment trap for existing detectors. Existing datasets lack pure synthesis fake news videos. Although directly prompting T2V models with descriptions of fake news videos can yield perfectly aligned samples, it reduces the fake news video detection (FNVD) to unimodal shortcuts and causes semantic-visual degeneration. To counter this, we formulate T2V-FNVD as a novel ternary classification task with three labels (real, cheap fake, and pure synthesis fake) and construct the first pure synthesis fake news video dataset (PS-FNVD). PS-FNVD includes fabricated events with aligned deception (Type 1) and true events with false visual provenance (Type 2), preventing models from exploiting unimodal shortcuts. Furthermore, we propose the Reasoning-guided T2V-FNVD (R-T2V) framework. Trained through conditioned rationale generation and supervised fine-tuning, R-T2V integrates high-level semantic logic with low-level physical generative traces to predict the ternary veracity label. Extensive experiments across 10 prevailing baselines show that R-T2V achieves the state-of-the-art performance, outperforming the second-best baseline by 12.20 percentage points in accuracy and 8.46 percentage points in macro F1.