Collaboration and Controversy Among Experts: Rumor Early Detection by Tuning a Comment Generator

• URL: http://arxiv.org/abs/2504.04076v1
• Date: Sat, 05 Apr 2025 06:21:01 GMT
• Title: Collaboration and Controversy Among Experts: Rumor Early Detection by Tuning a Comment Generator
• Authors: Bing Wang, Bingrui Zhao, Ximing Li, Changchun Li, Wanfu Gao, Shengsheng Wang,
• Abstract summary: We propose a new Rumor Early Detection (RED) framework named CAMERED.<n>We integrate a mixture-of-expert structure into a generative language model and present a novel routing network for expert collaboration.<n>Results show that CAMERED outperforms state-of-the-art RED baseline models and generation methods.
• Score: 20.957905317077046
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Over the past decade, social media platforms have been key in spreading rumors, leading to significant negative impacts. To counter this, the community has developed various Rumor Detection (RD) algorithms to automatically identify them using user comments as evidence. However, these RD methods often fail in the early stages of rumor propagation when only limited user comments are available, leading the community to focus on a more challenging topic named Rumor Early Detection (RED). Typically, existing RED methods learn from limited semantics in early comments. However, our preliminary experiment reveals that the RED models always perform best when the number of training and test comments is consistent and extensive. This inspires us to address the RED issue by generating more human-like comments to support this hypothesis. To implement this idea, we tune a comment generator by simulating expert collaboration and controversy and propose a new RED framework named CAMERED. Specifically, we integrate a mixture-of-expert structure into a generative language model and present a novel routing network for expert collaboration. Additionally, we synthesize a knowledgeable dataset and design an adversarial learning strategy to align the style of generated comments with real-world comments. We further integrate generated and original comments with a mutual controversy fusion module. Experimental results show that CAMERED outperforms state-of-the-art RED baseline models and generation methods, demonstrating its effectiveness.

Related papers

• Investigating and Improving Counter-Stereotypical Action Relation in Text-to-Image Diffusion Models [28.49695567630899]
  Text-to-image diffusion models consistently fail at generating counter-stereotypical action relationships. We discover this limitation stems from distributional biases rather than inherent model constraints.
  arXiv  Detail & Related papers  (2025-03-13T04:38:02Z)
• Generative Adversarial Reviews: When LLMs Become the Critic [1.2430809884830318]
  We introduce Generative Agent Reviewers (GAR), leveraging LLM-empowered agents to simulate faithful peer reviewers.<n>Central to this approach is a graph-based representation of manuscripts, condensing content and logically organizing information.<n>Our experiments demonstrate that GAR performs comparably to human reviewers in providing detailed feedback and predicting paper outcomes.
  arXiv  Detail & Related papers  (2024-12-09T06:58:17Z)
• Context Enhancement with Reconstruction as Sequence for Unified Unsupervised Anomaly Detection [68.74469657656822]
  Unsupervised anomaly detection (AD) aims to train robust detection models using only normal samples. Recent research focuses on a unified unsupervised AD setting in which only one model is trained for all classes. We introduce a novel Reconstruction as Sequence (RAS) method, which enhances the contextual correspondence during feature reconstruction.
  arXiv  Detail & Related papers  (2024-09-10T07:37:58Z)
• SHINE: Saliency-aware HIerarchical NEgative Ranking for Compositional Temporal Grounding [52.98133831401225]
  Temporal grounding, also known as video moment retrieval, aims at locating video segments corresponding to a given query sentence. We propose a large language model-driven method for negative query construction, utilizing GPT-3.5-Turbo. We introduce a coarse-to-fine saliency ranking strategy, which encourages the model to learn the multi-granularity semantic relationships between videos and hierarchical negative queries.
  arXiv  Detail & Related papers  (2024-07-06T16:08:17Z)
• Let Silence Speak: Enhancing Fake News Detection with Generated Comments from Large Language Models [17.612043837566134]
  Comments could reflect users' opinions, stances, and emotions and models deepen their understanding of fake news. Due to exposure bias and users' different willingness to comment, it is not easy to obtain diverse comments in reality. We propose GenFEND, a generated feedback-enhanced detection framework, which generates comments by prompting LLMs with diverse user profiles.
  arXiv  Detail & Related papers  (2024-05-26T17:09:23Z)
• Promoting Constructive Deliberation: Reframing for Receptiveness [5.4346288442609945]
  We propose automatic reframing of disagreeing responses to signal receptiveness to a preceding comment. We identify six strategies for reframing. We automatically reframe replies to comments according to each strategy, using a Reddit dataset. We find that the replies generated with our framework are perceived to be significantly more receptive than the original replies and a generic receptiveness baseline.
  arXiv  Detail & Related papers  (2024-05-23T21:35:22Z)
• RELIC: Investigating Large Language Model Responses using Self-Consistency [58.63436505595177]
  Large Language Models (LLMs) are notorious for blending fact with fiction and generating non-factual content, known as hallucinations. We propose an interactive system that helps users gain insight into the reliability of the generated text.
  arXiv  Detail & Related papers  (2023-11-28T14:55:52Z)
• ViCo: Engaging Video Comment Generation with Human Preference Rewards [68.50351391812723]
  We propose ViCo with three novel designs to tackle the challenges for generating engaging Video Comments. To quantify the engagement of comments, we utilize the number of "likes" each comment receives as a proxy of human preference. To automatically evaluate the engagement of comments, we train a reward model to align its judgment to the above proxy.
  arXiv  Detail & Related papers  (2023-08-22T04:01:01Z)
• FLIRT: Feedback Loop In-context Red Teaming [79.63896510559357]
  We propose an automatic red teaming framework that evaluates a given black-box model and exposes its vulnerabilities. Our framework uses in-context learning in a feedback loop to red team models and trigger them into unsafe content generation.
  arXiv  Detail & Related papers  (2023-08-08T14:03:08Z)
• Shilling Black-box Review-based Recommender Systems through Fake Review Generation [20.162253355141893]
  Review-Based Recommender Systems (RBRS) have attracted increasing research interest due to their ability to alleviate well-known cold-start problems. We argue that such a reliance on reviews may instead expose systems to the risk of being shilled. We propose the first generation-based model for shilling attacks against RBRSs.
  arXiv  Detail & Related papers  (2023-06-27T12:32:36Z)
• Verifying the Robustness of Automatic Credibility Assessment [50.55687778699995]
  We show that meaning-preserving changes in input text can mislead the models. We also introduce BODEGA: a benchmark for testing both victim models and attack methods on misinformation detection tasks. Our experimental results show that modern large language models are often more vulnerable to attacks than previous, smaller solutions.
  arXiv  Detail & Related papers  (2023-03-14T16:11:47Z)
• SIFN: A Sentiment-aware Interactive Fusion Network for Review-based Item Recommendation [48.1799451277808]
  We propose a Sentiment-aware Interactive Fusion Network (SIFN) for review-based item recommendation. We first encode user/item reviews via BERT and propose a light-weighted sentiment learner to extract semantic features of each review. Then, we propose a sentiment prediction task that guides the sentiment learner to extract sentiment-aware features via explicit sentiment labels.
  arXiv  Detail & Related papers  (2021-08-18T08:04:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.