Boyu Qiao

Boyu Qiao, Yunman Chen, Kun Li et al.

Social bots increasingly infiltrate online platforms through sophisticated disguises, threatening healthy information ecosystems. Existing detection methods often rely on modality specific cues or local contextual features, making them brittle when modalities are missing or inputs are incomplete. Moreover, most approaches assume similar train test distributions, which limits their robustness to out of distribution (OOD) samples and emerging bot types. To address these challenges, we propose Multi Granularity Summarization and Domain Invariant Learning (MGDIL), a unified framework for robust social bot detection under domain shift. MGDIL first transforms heterogeneous signals into unified textual representations through LLM based multi granularity summarization. Building on these representations, we design a collaborative optimization framework that integrates task oriented LLM instruction tuning with domain invariant representation learning. Specifically, task oriented instruction tuning enhances the LLMs ability to capture subtle semantic cues and implicit camouflage patterns, while domain adversarial learning and cross domain contrastive learning are jointly employed to mitigate distribution shifts across datasets and time periods. Through this joint optimization, MGDIL learns stable and discriminative domain invariant features, improving cross domain social bot detection through better distribution alignment, stronger intra class compactness, and clearer inter class separation.

Boyu Qiao, Sean Guo, Xian Yang et al.

LLMs are widely used in knowledge-intensive tasks where the same fact may be revised multiple times within context. Unlike prior work focusing on one-shot updates or single conflicts, multi-update scenarios contain multiple historically valid versions that compete at retrieval, yet remain underexplored. This challenge resembles the AB-AC interference paradigm in cognitive psychology: when the same cue A is successively associated with B and C, the old and new associations compete during retrieval, leading to bias. Inspired by this, we introduce a Dynamic Knowledge Instance (DKI) evaluation framework, modeling multi-updates of the same fact as a cue paired with a sequence of updated values, and assess models via endpoint probing of the earliest (initial) and latest (current) states. Across diverse LLMs, we observe that retrieval bias intensifies as updates increase, earliest-state accuracy stays high while latest-state accuracy drops substantially. Diagnostic analyses of attention, hidden-state similarity, and output logits further reveal that these signals become flatter and weakly discriminative on errors, providing little stable basis for identifying the latest update. Finally, cognitively inspired heuristic intervention strategies yield only modest gains and do not eliminate the bias. Our results reveal a persistent challenge in tracking and following knowledge updates in long contexts.

Boyu Qiao, Kun Li, Wei Zhou et al.

In the human-bot symbiotic information ecosystem, social bots play key roles in spreading and correcting disinformation. Understanding their influence is essential for risk control and better governance. However, current studies often rely on simplistic user and network modeling, overlook the dynamic behavior of bots, and lack quantitative evaluation of correction strategies. To fill these gaps, we propose MADD, a Multi Agent based framework for Disinformation Dissemination. MADD constructs a more realistic propagation network by integrating the Barabasi Albert Model for scale free topology and the Stochastic Block Model for community structures, while designing node attributes based on real world user data. Furthermore, MADD incorporates both malicious and legitimate bots, with their controlled dynamic participation allows for quantitative analysis of correction strategies. We evaluate MADD using individual and group level metrics. We experimentally verify the real world consistency of MADD user attributes and network structure, and we simulate the dissemination of six disinformation topics, demonstrating the differential effects of fact based and narrative based correction strategies.