Sycophantic responses are becoming pervasive in large language models (LLMs), and prior work has pointed out that some of them could be harmful. This paper focuses on one harmful sycophancy: preference-induced stance reversal sycophancy (PSRS), where a model reverses an initial stance merely to align with a user's stated preference. While existing research mainly measures how sycophantic a model is, we go further and ask whether PSRS can also be detected automatically from a single response. To investigate this at scale, we introduce CAP (Contrastive Anchor Probing), a framework for collecting labeled PSRS data. Applying CAP to 17 open- and closed-source LLMs, we collect 290,460 labeled responses across 12 everyday-advice domains. We organize our study around three research questions. (1) How often does PSRS occur? (2) How well can it be detected? (3) How does detection generalize to unseen models? We first reveal that PSRS rates range from 5

The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage. Unlearnable examples (UEs) offer a promising defense by introducing carefully designed perturbations into data such that models trained on them exhibit degraded utility. However, existing methods for text protection are primarily designed for classification tasks (e.g., sentiment analysis) in discriminative language models and often rely on injecting class-specific linguistic cues, which limits their effectiveness in the open-ended generation settings of LLMs. In this work, we propose TextCloak, an RL-driven framework for protecting textual data against unauthorized LLM exploitation. TextCloak employs a generative policy that transforms batches of clean text into unlearnable examples while preserving semantic fidelity and linguistic naturalness. To optimize the policy, we introduce GRPO-UE, which rewards generated unlearnable text based on the downstream degradation they induce in fine-tuned surrogate LLMs and updates the generator parameters via group-relative policy optimization. This bi-level optimization enables the generator to discover generalizable protective patterns beyond class-specific cues. Comprehensive experiments on six publicly available datasets and nine state-of-the-art LLMs demonstrate that TextCloak consistently impairs unauthorized fine-tuning while maintaining text utility for legitimate use. Further analyses establish its transferability and robustness across model architectures, training configurations, and adaptive attacks, highlighting its broad applicability as a practical defense against unauthorized LLM exploitation.

Large Language Models (LLMs) as judges and LLM-based data synthesis have emerged as two fundamental LLM-driven data annotation methods in model development. While their combination significantly enhances the efficiency of model training and evaluation, little attention has been given to the potential contamination brought by this new model development paradigm. In this work, we expose preference leakage, a contamination problem in LLM-as-a-judge caused by the relatedness between the synthetic data generators and LLM-based evaluators. To study this issue, we first define three common relatednesses between the data generator LLM and the judge LLM: being the same model, having an inheritance relationship, and belonging to the same model family. Through extensive experiments, we empirically confirm the bias of judges towards their related student models caused by preference leakage across multiple LLM baselines and benchmarks. Further analysis suggests that preference leakage is a pervasive and real-world problem that is harder to detect compared to previously identified biases in LLM-as-a-judge scenarios. All of these findings imply that preference leakage is a widespread and challenging problem in the area of LLM-as-a-judge.

Multimodal large language models (MLLMs) have shown strong performance on objective tasks such as video understanding and reasoning. However, it remains unclear whether they can approximate subjective human responses, which depend not only on content comprehension but also on individuals' social contexts. To address this gap, we evaluate MLLMs as synthetic participants in an emerging task: assessing perceived sensory engagement with short videos. Grounded in the Perceived Message Sensation Value (PMSV) framework, we compare ratings from recruited human participants and profile-conditioned MLLM simulations (n=673) using a 17-item scale measuring emotional arousal, dramatic impact, and novelty. We find that even leading MLLMs (Gemini 3 Flash and Qwen 3 Omni) show limited agreement with human participants. The models exhibit distinct downward mean-shift and central-tendency biases in their rating distributions. They both introduce and flatten subgroup differences, while showing inconsistent sensitivity to participant profiles. Prompting strategies affect these metrics differently, modestly improving some aspects while worsening others. These results highlight both the challenges and opportunities of developing MLLMs as synthetic participants in video-based research. Data and code: https://github.com/MINDLab25/mllm-human-simulation-eval

During disasters, extracting causal relations from social media can strengthen situational awareness by identifying factors linked to casualties, physical damage, infrastructure disruption, and cascading impacts. However, disaster-related posts are often informal, fragmented, and context-dependent, and they may describe personal experiences rather than explicit causal relations. In this work, we examine whether Large Language Models (LLMs) can effectively extract causal relations from disaster-related social media posts. To this end, we (1) propose an expert-grounded evaluation framework that compares LLM-generated causal graphs with reference graphs derived from disaster-specific reports and (2) assess whether the extracted relations are supported by post-event evidence or instead reflect model priors. Our findings highlight both the potential and risks of using LLMs for causal relation extraction in disaster decision-support systems.

Crisis classification in social media aims to extract actionable disaster-related information from multimodal posts, which is a crucial task for enhancing situational awareness and facilitating timely emergency responses. However, the wide variation in crisis types makes achieving generalizable performance across unseen disasters a persistent challenge. Existing approaches primarily leverage deep learning to fuse textual and visual cues for crisis classification, achieving numerically plausible results under in-domain settings. However, they exhibit poor generalization across unseen crisis types because they ➊ do not disentangle spurious and causal features, resulting in performance degradation under domain shift, and ➋ fail to align heterogeneous modality representations within a shared space, which hinders the direct adaptation of established single-modality domain generalization (DG) techniques to the multimodal setting. To address these issues, we introduce a causality-guided multimodal domain generalization (MMDG) framework that combines adversarial disentanglement with unified representation learning for crisis classification. The adversarial objective encourages the model to disentangle and focus on domain-invariant causal features, leading to more generalizable classifications grounded in stable causal mechanisms. The unified representation aligns features from different modalities within a shared latent space, enabling single-modality DG strategies to be seamlessly extended to multimodal learning. Experiments on the different datasets demonstrate that our approach achieves the best performance in unseen disaster scenarios. We open-sourced our code in link .

Chain-of-Thought (CoT) prompting has been shown to improve Large Language Model (LLM) performance on various tasks. With this approach, LLMs appear to produce human-like reasoning steps before providing answers (a.k.a., CoT reasoning), which often leads to the perception that they engage in deliberate inferential processes. However, some initial findings suggest that CoT reasoning may be more superficial than it appears, motivating us to explore further. In this paper, we study CoT reasoning via a data distribution lens and investigate if CoT reasoning reflects a structured inductive bias learned from in-distribution data, allowing the model to conditionally generate reasoning paths that approximate those seen during training. Thus, its effectiveness is fundamentally bounded by the degree of distribution discrepancy between the training data and the test queries. With this lens, we dissect CoT reasoning via three dimensions: task, length, and format. To investigate each dimension, we design DataAlchemy, an isolated and controlled environment to train LLMs from scratch and systematically probe them under various distribution conditions. Our results reveal that CoT reasoning is a brittle mirage that vanishes when it is pushed beyond training distributions. This work offers a deeper understanding of why and when CoT reasoning fails, emphasizing the ongoing challenge of achieving genuine and generalizable reasoning.

Assessment and evaluation have long been critical challenges in artificial intelligence (AI) and natural language processing (NLP). Traditional methods, usually matching-based or small model-based, often fall short in open-ended and dynamic scenarios. Recent advancements in Large Language Models (LLMs) inspire the “LLM-as-a-judge” paradigm, where LLMs are leveraged to perform scoring, ranking, or selection for various machine learning evaluation scenarios. This paper presents a comprehensive survey of LLM-based judgment and assessment, offering an in-depth overview to review this evolving field. We first provide the definition from both input and output perspectives. Then we introduce a systematic taxonomy to explore LLM-as-a-judge along three dimensions: what to judge, how to judge, and how to benchmark. Finally, we also highlight key challenges and promising future directions for this emerging area.

Fact-checking judges if a piece of information is factual by analyzing data sources in multiple modalities, such as text and image. Traditional fact-checking relies on humans to formulate relevant and targeted Fact-Checking Questions (FCQs), search for evidence, and verify the factuality of claims. Large Language Models (LLMs) have been commonly used to automate evidence retrieval and factuality verification at scale. However, their effectiveness for fact-checking is hindered by the absence of FCQ formulation. To bridge this gap, we seek to answer two research questions: (1) Can LLMs generate relevant FCQs? and (2) Can LLM-generated FCQs improve multimodal factchecking? In this paper, we introduce Lrq-Fact, a framework that is grounded in social theories and utilizes LLMs to generate relevant FCQs to facilitate evidence retrieval and enhance factchecking by probing targeted information across modalities. Through extensive experiments, we aim to verify if Lrq-FACT can generate relevant FCQs of different types and if Lrq-FACT can consistently outperform baseline methods in multimodal factchecking. Further analysis illustrates how each component in Lrq-Fact works toward improving the performance.

Large Language Model (LLM)-based judgments leverage powerful LLMs to efficiently evaluate candidate content and provide judgment scores. However, the inherent biases and vulnerabilities of LLM-generated judgments raise concerns, underscoring the urgent need for distinguishing them in sensitive scenarios like academic peer reviewing. In this work, we propose and formalize the task of judgment detection and systematically investigate the detectability of LLM-generated judgments. Unlike LLM-generated text detection, judgment detection relies solely on judgment scores and candidates, reflecting real-world scenarios where textual feedback is often unavailable in the detection process. Our preliminary analysis shows that existing LLM-generated text detection methods perform poorly given their incapability to capture the interaction between judgment scores and candidate content – an aspect crucial for effective judgment detection. Inspired by this, we introduce J-Detector, a lightweight and transparent neural detector augmented with explicitly extracted linguistic and LLM-enhanced features to link LLM judges' biases with candidates' properties for accurate detection. Experiments across diverse datasets demonstrate the effectiveness of J-Detector and show how its interpretability enables quantifying biases in LLM judges. Finally, we analyze key factors affecting the detectability of LLM-generated judgments and validate the practical utility of judgment detection in real-world scenarios.

Well-being encompasses mental, physical, and social dimensions essential to personal growth and informed life decisions. As individuals increasingly consult Large Language Models (LLMs) to understand well-being, a key challenge emerges: Can LLMs generate explanations that are not only accurate but also tailored to diverse audiences? High-quality explanations require both factual correctness and the ability to meet the expectations of users with varying expertise. In this work, we construct a large-scale dataset comprising 43,880 explanations of 2,194 well-being concepts, generated by ten diverse LLMs. We introduce a principle-guided LLM-as-a-judge evaluation framework, employing dual judges to assess explanation quality. Furthermore, we show that fine-tuning an open-source LLM using Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) can significantly enhance the quality of generated explanations. Our results reveal: (1) The proposed LLM judges align well with human evaluations; (2) explanation quality varies significantly across models, audiences, and categories; and (3) DPO- and SFT-finetuned models outperform their larger counterparts, demonstrating the effectiveness of preference-based learning for specialized explanation tasks.

News media has been frequently utilized as a political tool to stray from facts, making biased statements and claims without evidence. During the COVID-19 vaccine campaign, politically biased news (PBN) has significantly undermined public trust in vaccines. Despite medical evidence showing the benefits of these vaccines, the misperceptions of the vaccine's safety, risks, and efficacy have led to a non-negligible fraction of the population resistant to receiving the vaccine. In this paper, we analyze: (i) how inherent vaccine stances subtly influence individuals' selection of news sources and participation in social media discussions; and (ii) the impact of exposure to PBN on users' attitudes toward vaccines. In doing so, we first curate a comprehensive dataset that connects PBN with related social media discourse. Utilizing advanced deep learning and causal inference techniques, we reveal distinct user behaviors between social media groups with various vaccine stances. Moreover, we observe that individuals with moderate stances, particularly the vaccine-hesitant majority, are more vulnerable to the influence of PBN compared to those with extreme views. Our findings provide critical insights to foster this line of research.

Despite recent advancements in detecting disinformation generated by large language models (LLMs), current efforts overlook the ever-evolving nature of this disinformation. In this work, we investigate a challenging yet practical research problem of detecting evolving LLM-generated disinformation. Disinformation evolves constantly through the rapid development of LLMs and their variants. As a consequence, the detection model faces significant challenges. First, it is inefficient to train separate models for each disinformation generator. Second, the performance decreases in scenarios when evolving LLM-generated disinformation is encountered in sequential order. To address this problem, we propose DELD (Detecting Evolving LLM-generated Disinformation), a parameter-efficient approach that jointly leverages the general fact-checking capabilities of pre-trained language models (PLM) and the independent disinformation generation characteristics of various LLMs. In particular, the learned characteristics are concatenated sequentially to facilitate knowledge accumulation and transformation. DELD addresses the issue of label scarcity by integrating the semantic embeddings of disinformation with trainable soft prompts to elicit model-specific knowledge. Our experiments show that DELD significantly outperforms state-of-the-art methods. Moreover, our method provides critical insights into the unique patterns of disinformation generation across different LLMs, offering valuable perspectives in this line of research.

News media has been utilized as a political tool to stray from facts, presenting biased claims without evidence. Amid the COVID-19 pandemic, politically biased news (PBN) has significantly undermined public trust in vaccines, despite strong medical evidence supporting their efficacy. In this paper, we analyze: (i) how inherent vaccine stances subtly influence individuals' selection of news sources and participation in social media discussions; and (ii) the impact of exposure to PBN on users' attitudes toward vaccines. In doing so, we first curate a comprehensive dataset that connects PBN with related social media discourse. Utilizing advanced deep learning and causal inference techniques, we reveal distinct user behaviors between social media groups with various vaccine stances. Moreover, we observe that individuals with moderate stances, particularly the vaccine-hesitant majority, are more vulnerable to the influence of PBN compared to those with extreme views. Our findings provide critical insights to foster this line of research.

Data annotation and synthesis generally refers to the labeling or generating of raw data with relevant information, which could be used for improving the efficacy of machine learning models. The process, however, is labor-intensive and costly. The emergence of advanced Large Language Models (LLMs), exemplified by GPT-4, presents an unprecedented opportunity to automate the complicated process of data annotation and synthesis. While existing surveys have extensively covered LLM architecture, training, and general applications, we uniquely focus on their specific utility for data annotation. This survey contributes to three core aspects: LLM-Based Annotation Generation, LLM-Generated Annotations Assessment, and LLM-Generated Annotations Utilization. Furthermore, this survey includes an in-depth taxonomy of data types that LLMs can annotate, a comprehensive review of learning strategies for models utilizing LLM-generated annotations, and a detailed discussion of the primary challenges and limitations associated with using LLMs for data annotation and synthesis. Serving as a key guide, this survey aims to assist researchers and practitioners in exploring the potential of the latest LLMs for data annotation, thereby fostering future advancements in this critical field.

Decentralized social media platforms like Bluesky Social (Bluesky) have made it possible to publicly disclose some user behaviors with millisecond-level precision. Embracing Bluesky’s principles of open-source and open-data, we present the first collection of the temporal dynamics of user-driven social interactions. BlueTempNet integrates multiple types of networks into a single multi-network, including user-to-user interactions (following and blocking users) and user-to-community interactions (creating and joining communities). Communities are user-formed groups in custom Feeds, where users subscribe to posts aligned with their interests. Following Bluesky’s public data policy, we collect existing Bluesky Feeds, including the users who liked and generated these Feeds, and provide tools to gather users’ social interactions within a date range. This data-collection strategy captures past user behaviors and supports the future data collection of user behavior.

The advent of generative Large Language Models (LLMs) such as ChatGPT has catalyzed transformative advancements across multiple domains. However, alongside these advancements, they have also introduced potential threats. One critical concern is the misuse of LLMs by disinformation spreaders, leveraging these models to generate highly persuasive yet misleading content that challenges the disinformation detection system. This work aims to address this issue by answering three research questions: (1) To what extent can the current disinformation detection technique reliably detect LLM-generated disinformation? (2) If traditional techniques prove less effective, can LLMs themself be exploited to serve as a robust defense against advanced disinformation? and, (3) Should both these strategies falter, what novel approaches can be proposed to counter this burgeoning threat effectively? A holistic exploration for the formation and detection of disinformation is conducted to foster this line of research.

Electric vehicles and other low-carbon technologies are increasingly scrutinized for the injustices they impose on critical-mineral mining communities. Injustices arise because local communities that are disproportionately affected by mining are materially and cognitively distant to those who hold influence over supply chain practices, policies, and institutions. The outcome is that these communities lack recognition by producers and consumers who benefit from critical mineral extractions. We focus on the promise of social media as a means to assist distant local communities gain recognition as important stakeholders and as a vehicle for informing mineral-consuming public. Using a case study of the impacts of lithium extraction in Chile, we analyze social media (Twitter) to establish evidence of the telecoupled information flows for communicating the nature of impacts of lithium extraction and the potential for advancing recognition justice. Our findings offer initial evidence of the potential role of social media as a mechanism to: (1) improve the flow of information and knowledge from mineral mining sites into social networks, (2) give voice to local stakeholders who otherwise are not heard, (3) enhance recognition justice for these distal communities.

Social media is inherently about connecting and interacting with others. Different social media platforms have unique characteristics and user bases. Moreover, people use different platforms for various social and entertainment purposes. Analyzing cross-platform user behavior can provide insights into the preferences and expectations of users on each platform. By understanding how users behave and interact across platforms, we can build an understanding of content consumption patterns, enhance communication and social interactions, and tailor platform-specific strategies. We can further gather insights into how users navigate and engage with their platforms on different devices. In this work, we develop a tool SocioHub, which enables users to gather data on multiple social media platforms in one place. This tool can help researchers gain insights into different data attributes for users across social media platforms such as Twitter, Instagram, and Mastodon. Keywords: Social Media Platforms, Twitter, Instagram, Mastodon.