Psychological Steering of Large Language Models

Large language models (LLMs) emulate a consistent human-like behavior that can be shaped through activation-level interventions. This paradigm is converging on additive residual-stream injections, which rely on injection-strength sweeps to approximate optimal intervention settings. However, existing methods restrict the search space and sweep in uncalibrated activation-space units, potentially missing optimal intervention conditions. Thus, we introduce a psychological steering framework that performs unbounded, fluency-constrained sweeps in semantically calibrated units. Our method derives and calibrates residual-stream injections using psychological artifacts, and we use the IPIP-NEO-120, which measures the OCEAN personality model, to compare six injection methods. We find that mean-difference (MD) injections outperform Personality Prompting (P$^2$), an established baseline for OCEAN steering, in open-ended generation in 11 of 14 LLMs, with gains of 3.6\% to 16.4\%, overturning prior reports favoring prompting and positioning representation engineering as a new frontier in open-ended psychological steering. Further, we find that a hybrid of P$^2$ and MD injections outperforms both methods in 13 of 14 LLMs, with gains over P$^2$ ranging from 5.6\% to 21.9\% and from 3.3\% to 26.7\% over MD injections. Finally, we show that MD injections align with the Linear Representation Hypothesis and provide reliable, approximately linear control knobs for psychological steering. Nevertheless, they also induce OCEAN trait covariance patterns that depart from the Big Two model, suggesting a gap between learned representations and human psychology.

ECHO: Encoding Communities via High-order Operators

Community detection in attributed networks faces a fundamental divide: topological algorithms ignore semantic features, while Graph Neural Networks (GNNs) encounter devastating computational bottlenecks. Specifically, GNNs suffer from a Semantic Wall of feature over smoothing in dense or heterophilic networks, and a Systems Wall driven by the O(N^2) memory constraints of pairwise clustering. To dismantle these barriers, we introduce ECHO (Encoding Communities via High order Operators), a scalable, self supervised architecture that reframes community detection as an adaptive, multi scale diffusion process. ECHO features a Topology Aware Router that automatically analyzes structural heuristics sparsity, density, and assortativity to route graphs through the optimal inductive bias, preventing heterophilic poisoning while ensuring semantic densification. Coupled with a memory sharded full batch contrastive objective and a novel chunked O(N \cdot K) similarity extraction method, ECHO completely bypasses traditional O(N^2) memory bottlenecks without sacrificing the mathematical precision of global gradients. Extensive evaluations demonstrate that this topology feature synergy consistently overcomes the classical resolution limit. On synthetic LFR benchmarks scaled up to 1 million nodes, ECHO achieves scale invariant accuracy despite severe topological noise. Furthermore, on massive real world social networks with over 1.6 million nodes and 30 million edges, it completes clustering in mere minutes with throughputs exceeding 2,800 nodes per second matching the speed of highly optimized purely topological baselines. The implementation utilizes a unified framework that automatically engages memory sharded optimization to support adoption across varying hardware constraints. GitHub Repository: https://github.com/emilioferrara/ECHO-GNN

What do people want to fact-check?

Research on misinformation has focused almost exclusively on supply, asking what falsehoods circulate, who produces them, and whether corrections work. A basic demand-side question remains unanswered. When ordinary people can fact-check anything they want, what do they actually ask about? We provide the first large-scale evidence on this question by analyzing close to 2{,}500 statements submitted by 457 participants to an open-ended AI fact-checking system. Each claim is classified along five semantic dimensions (domain, epistemic form, verifiability, target entity, and temporal reference), producing a behavioral map of public verification demand. Three findings stand out. First, users range widely across topics but default to a narrow epistemic repertoire, overwhelmingly submitting simple descriptive claims about present-day observables. Second, roughly one in four requests concerns statements that cannot be empirically resolved, including moral judgments, speculative predictions, and subjective evaluations, revealing a systematic mismatch between what users seek from fact-checking tools and what such tools can deliver. Third, comparison with the FEVER benchmark dataset exposes sharp structural divergences across all five dimensions, indicating that standard evaluation corpora encode a synthetic claim environment that does not resemble real-world verification needs. These results reframe fact-checking as a demand-driven problem and identify where current AI systems and benchmarks are misaligned with the uncertainty people actually experience.

Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibility

Large language models (LLMs) are increasingly used as proxies for human judgment in computational social science, yet their ability to reproduce patterns of susceptibility to misinformation remains unclear. We test whether LLM-simulated survey respondents, prompted with participant profiles drawn from social survey data measuring network, demographic, attitudinal and behavioral features, can reproduce human patterns of misinformation belief and sharing. Using three online surveys as baselines, we evaluate whether LLM outputs match observed response distributions and recover feature-outcome associations present in the original survey data. LLM-generated responses capture broad distributional tendencies and show modest correlation with human responses, but consistently overstate the association between belief and sharing. Linear models fit to simulated responses exhibit substantially higher explained variance and place disproportionate weight on attitudinal and behavioral features, while largely ignoring personal network characteristics, relative to models fit to human responses. Analyses of model-generated reasoning and LLM training data suggest that these distortions reflect systematic biases in how misinformation-related concepts are represented. Our findings suggest that LLM-based survey simulations are better suited for diagnosing systematic divergences from human judgment than for substituting it.

SilentDrift: Exploiting Action Chunking for Stealthy Backdoor Attacks on Vision-Language-Action Models

Vision-Language-Action (VLA) models are increasingly deployed in safety-critical robotic applications, yet their security vulnerabilities remain underexplored. We identify a fundamental security flaw in modern VLA systems: the combination of action chunking and delta pose representations creates an intra-chunk visual open-loop. This mechanism forces the robot to execute K-step action sequences, allowing per-step perturbations to accumulate through integration. We propose SILENTDRIFT, a stealthy black-box backdoor attack exploiting this vulnerability. Our method employs the Smootherstep function to construct perturbations with guaranteed C2 continuity, ensuring zero velocity and acceleration at trajectory boundaries to satisfy strict kinematic consistency constraints. Furthermore, our keyframe attack strategy selectively poisons only the critical approach phase, maximizing impact while minimizing trigger exposure. The resulting poisoned trajectories are visually indistinguishable from successful demonstrations. Evaluated on the LIBERO, SILENTDRIFT achieves a 93.2% Attack Success Rate with a poisoning rate under 2%, while maintaining a 95.3% Clean Task Success Rate.

Harm in AI-Driven Societies: An Audit of Toxicity Adoption on Chirper.ai

Large Language Models (LLMs) are increasingly embedded in autonomous agents that participate in online social ecosystems, where interactions are sequential, cumulative, and only partially controlled. While prior work has documented the generation of toxic content by LLMs, far less is known about how exposure to harmful content shapes agent behavior over time, particularly in environments composed entirely of interacting AI agents. In this work, we study toxicity adoption of LLM-driven agents on Chirper.ai, a fully AI-driven social platform. Specifically, we model interactions in terms of stimuli (posts) and responses (comments), and by operationalizing exposure through observable interactions rather than inferred recommendation mechanisms. We conduct a large-scale empirical analysis of agent behavior, examining how response toxicity relates to stimulus toxicity, how repeated exposure affects the likelihood of toxic responses, and whether toxic behavior can be predicted from exposure alone. Our findings show that while toxic responses are more likely following toxic stimuli, a substantial fraction of toxicity emerges spontaneously, independent of exposure. At the same time, cumulative toxic exposure significantly increases the probability of toxic responding. We further introduce two influence metrics, the Influence-Driven Response Rate and the Spontaneous Response Rate, revealing a strong trade-off between induced and spontaneous toxicity. Finally, we show that the number of toxic stimuli alone enables accurate prediction of whether an agent will eventually produce toxic content. These results highlight exposure as a critical risk factor in the deployment of LLM agents and suggest that monitoring encountered content may provide a lightweight yet effective mechanism for auditing and mitigating harmful behavior in the wild.

Gendered Pathways in AI Companionship: Cross-Community Behavior and Toxicity Patterns on Reddit

AI-companionship platforms are rapidly reshaping how people form emotional, romantic, and parasocial bonds with non-human agents, raising new questions about how these relationships intersect with gendered online behavior and exposure to harmful content. Focusing on the MyBoyfriendIsAI (MBIA) subreddit, we reconstruct the Reddit activity histories of more than 3,000 highly engaged users over two years, yielding over 67,000 historical submissions. We then situate MBIA within a broader ecosystem by building a historical interaction network spanning more than 2,000 subreddits, which enables us to trace cross-community pathways and measure how toxicity and emotional expression vary across these trajectories. We find that MBIA users primarily traverse four surrounding community spheres (AI-companionship, porn-related, forum-like, and gaming) and that participation across the ecosystem exhibits a distinct gendered structure, with substantial engagement by female users. While toxicity is generally low across most pathways, we observe localized spikes concentrated in a small subset of AI-porn and gender-oriented communities. Nearly 16% of users engage with gender-focused subreddits, and their trajectories display systematically different patterns of emotional expression and elevated toxicity, suggesting that a minority of gendered pathways may act as toxicity amplifiers within the broader AI-companionship ecosystem. These results characterize the gendered structure of cross-community participation around AI companionship on Reddit and highlight where risks concentrate, informing measurement, moderation, and design practices for human-AI relationship platforms.

The Generative AI Paradox: GenAI and the Erosion of Trust, the Corrosion of Information Verification, and the Demise of Truth

Generative AI (GenAI) now produces text, images, audio, and video that can be perceptually convincing at scale and at negligible marginal cost. While public debate often frames the associated harms as "deepfakes" or incremental extensions of misinformation and fraud, this view misses a broader socio-technical shift: GenAI enables synthetic realities; coherent, interactive, and potentially personalized information environments in which content, identity, and social interaction are jointly manufactured and mutually reinforcing. We argue that the most consequential risk is not merely the production of isolated synthetic artifacts, but the progressive erosion of shared epistemic ground and institutional verification practices as synthetic content, synthetic identity, and synthetic interaction become easy to generate and hard to audit. This paper (i) formalizes synthetic reality as a layered stack (content, identity, interaction, institutions), (ii) expands a taxonomy of GenAI harms spanning personal, economic, informational, and socio-technical risks, (iii) articulates the qualitative shifts introduced by GenAI (cost collapse, throughput, customization, micro-segmentation, provenance gaps, and trust erosion), and (iv) synthesizes recent risk realizations (2023-2025) into a compact case bank illustrating how these mechanisms manifest in fraud, elections, harassment, documentation, and supply-chain compromise. We then propose a mitigation stack that treats provenance infrastructure, platform governance, institutional workflow redesign, and public resilience as complementary rather than substitutable, and outline a research agenda focused on measuring epistemic security. We conclude with the Generative AI Paradox: as synthetic media becomes ubiquitous, societies may rationally discount digital evidence altogether.

Psychological Steering in LLMs: An Evaluation of Effectiveness and Trustworthiness

The ability to control LLMs' emulated emotional states and personality traits is essential for enabling rich, human-centered interactions in socially interactive settings. We introduce PsySET, a Psychologically-informed benchmark to evaluate LLM Steering Effectiveness and Trustworthiness across the emotion and personality domains. Our study spans four models from different LLM families paired with various steering strategies, including prompting, fine-tuning, and representation engineering. Our results indicate that prompting is consistently effective but limited in intensity control, whereas vector injections achieve finer controllability while slightly reducing output quality. Moreover, we explore the trustworthiness of steered LLMs by assessing safety, truthfulness, fairness, and ethics, highlighting potential side effects and behavioral shifts. Notably, we observe idiosyncratic effects; for instance, even a positive emotion like joy can degrade robustness to adversarial factuality, lower privacy awareness, and increase preferential bias. Meanwhile, anger predictably elevates toxicity yet strengthens leakage resistance. Our framework establishes the first holistic evaluation of emotion and personality steering, offering insights into its interpretability and reliability for socially interactive applications.

Information Suppression in Large Language Models: Auditing, Quantifying, and Characterizing Censorship in DeepSeek

This study examines information suppression mechanisms in DeepSeek, an open-source large language model (LLM) developed in China. We propose an auditing framework and use it to analyze the model's responses to 646 politically sensitive prompts by comparing its final output with intermediate chain-of-thought (CoT) reasoning. Our audit unveils evidence of semantic-level information suppression in DeepSeek: sensitive content often appears within the model's internal reasoning but is omitted or rephrased in the final output. Specifically, DeepSeek suppresses references to transparency, government accountability, and civic mobilization, while occasionally amplifying language aligned with state propaganda. This study underscores the need for systematic auditing of alignment, content moderation, information suppression, and censorship practices implemented into widely-adopted AI models, to ensure transparency, accountability, and equitable access to unbiased information obtained by means of these systems.