We introduce and study artificial impressions--patterns in LLMs' internal representations of prompts that resemble human impressions and stereotypes based on language. We fit linear probes on generated prompts to predict impressions according to the two-dimensional Stereotype Content Model (SCM). Using these probes, we study the relationship between impressions and downstream model behavior as well as prompt features that may inform such impressions. We find that LLMs inconsistently report impressions when prompted, but also that impressions are more consistently linearly decodable from their hidden representations. Additionally, we show that artificial impressions of prompts are predictive of the quality and use of hedging in model responses. We also investigate how particular content, stylistic, and dialectal features in prompts impact LLM impressions.

Interprofessional education has long relied on case studies and the use of standardized patients to support teamwork, communication, and related collaborative competencies among healthcare professionals. However, traditional approaches are often limited by cost, scalability, and inability to mimic the dynamic complexity of real-world clinical scenarios. To address these challenges, we designed and developed AIMS (AI-Enhanced Immersive Multidisciplinary Simulations), a virtual simulation that integrates a large language model (Gemini-2.5-Flash), a Unity-based virtual environment engine, and a character creation pipeline to support synchronized, multimodal interactions between the user and the virtual patient. AIMS was designed to enhance collaborative clinical reasoning and health promotion competencies among students from pharmacy, medicine, nursing, and social work. A formal usability testing session was conducted which participants assumed professional roles on a healthcare team and engaged in a mix of scripted and unscripted conversations. Participants explored the patient's symptoms, social context, and care needs. Usability issues were identified (e.g., audio routing, response latency) and used to guide subsequent refinements. Findings in general suggest that AIMS supports realistic, profession-specific and contextually appropriate conversations. We discussed both technical and pedagogical innovations of AIMS and concluded with future directions.

Large language models (LLMs) remain vulnerable to multi-turn jailbreaking attacks that exploit conversational context to bypass safety constraints gradually. These attacks target different harm categories (like malware generation, harassment, or fraud) through distinct conversational approaches (educational discussions, personal experiences, hypothetical scenarios). Existing multi-turn jailbreaking methods often rely on heuristic or ad hoc exploration strategies, providing limited insight into underlying model weaknesses. The relationship between conversation patterns and model vulnerabilities across harm categories remains poorly understood. We propose Pattern Enhanced Chain of Attack (PE-CoA), a framework of five conversation patterns to construct effective multi-turn jailbreaks through natural dialogue. Evaluating PE-CoA on twelve LLMs spanning ten harm categories, we achieve state-of-the-art performance, uncovering pattern-specific vulnerabilities and LLM behavioral characteristics: models exhibit distinct weakness profiles where robustness to one conversational pattern does not generalize to others, and model families share similar failure modes. These findings highlight limitations of safety training and indicate the need for pattern-aware defenses. Code available on: https://github.com/Ragib-Amin-Nihal/PE-CoA

As AI writing tools become widespread, we need to understand how both humans and machines evaluate literary style, a domain where objective standards are elusive and judgments are inherently subjective. We conducted controlled experiments using Raymond Queneau's Exercises in Style (1947) to measure attribution bias across evaluators. Study 1 compared human participants (N=556) and AI models (N=13) evaluating literary passages from Queneau versus GPT-4-generated versions under three conditions: blind, accurately labeled, and counterfactually labeled. Study 2 tested bias generalization across a 14$\times$14 matrix of AI evaluators and creators. Both studies revealed systematic pro-human attribution bias. Humans showed +13.7 percentage point (pp) bias (Cohen's h = 0.28, 95% CI: 0.21-0.34), while AI models showed +34.3 percentage point bias (h = 0.70, 95% CI: 0.65-0.76), a 2.5-fold stronger effect (P$<$0.001). Study 2 confirmed this bias operates across AI architectures (+25.8pp, 95% CI: 24.1-27.6%), demonstrating that AI systems systematically devalue creative content when labeled as "AI-generated" regardless of which AI created it. We also find that attribution labels cause evaluators to invert assessment criteria, with identical features receiving opposing evaluations based solely on perceived authorship. This suggests AI models have absorbed human cultural biases against artificial creativity during training. Our study represents the first controlled comparison of attribution bias between human and artificial evaluators in aesthetic judgment, revealing that AI systems not only replicate but amplify this human tendency.

When learning to code, students often develop misconceptions about various programming language concepts. These can not only lead to bugs or inefficient code, but also slow down the learning of related concepts. In this paper, we introduce McMining, the task of mining programming misconceptions from samples of code from a student. To enable the training and evaluation of McMining systems, we develop an extensible benchmark dataset of misconceptions together with a large set of code samples where these misconceptions are manifested. We then introduce two LLM-based McMiner approaches and through extensive evaluations show that models from the Gemini, Claude, and GPT families are effective at discovering misconceptions in student code.

Understanding learner emotions in online education is critical for improving engagement and personalized instruction. While prior work in emotion recognition has explored multimodal fusion and temporal modeling, existing methods often rely on static fusion strategies and assume that modality inputs are consistently reliable, which is rarely the case in real-world learning environments. We introduce Edu-EmotionNet, a novel framework that jointly models temporal emotion evolution and modality reliability for robust affect recognition. Our model incorporates three key components: a Cross-Modality Attention Alignment (CMAA) module for dynamic cross-modal context sharing, a Modality Importance Estimator (MIE) that assigns confidence-based weights to each modality at every time step, and a Temporal Feedback Loop (TFL) that leverages previous predictions to enforce temporal consistency. Evaluated on educational subsets of IEMOCAP and MOSEI, re-annotated for confusion, curiosity, boredom, and frustration, Edu-EmotionNet achieves state-of-the-art performance and demonstrates strong robustness to missing or noisy modalities. Visualizations confirm its ability to capture emotional transitions and adaptively prioritize reliable signals, making it well suited for deployment in real-time learning systems

We propose a Geometry-aware Policy Imitation (GPI) approach that rethinks imitation learning by treating demonstrations as geometric curves rather than collections of state-action samples. From these curves, GPI derives distance fields that give rise to two complementary control primitives: a progression flow that advances along expert trajectories and an attraction flow that corrects deviations. Their combination defines a controllable, non-parametric vector field that directly guides robot behavior. This formulation decouples metric learning from policy synthesis, enabling modular adaptation across low-dimensional robot states and high-dimensional perceptual inputs. GPI naturally supports multimodality by preserving distinct demonstrations as separate models and allows efficient composition of new demonstrations through simple additions to the distance field. We evaluate GPI in simulation and on real robots across diverse tasks. Experiments show that GPI achieves higher success rates than diffusion-based policies while running 20 times faster, requiring less memory, and remaining robust to perturbations. These results establish GPI as an efficient, interpretable, and scalable alternative to generative approaches for robotic imitation learning. Project website: https://yimingli1998.github.io/projects/GPI/

Accurate recognition and re-identification of individual animals is essential for successful wildlife population monitoring. Traditional methods, such as leg banding of birds, are time consuming and invasive. Recent progress in artificial intelligence, particularly computer vision, offers encouraging solutions for smart conservation and efficient automation. This study presents a unique pipeline for extracting high-quality key frames from videos of kākā (Nestor meridionalis), a threatened forest-dwelling parrot in New Zealand. Key frame extraction is well-studied in person re-identification, however, its application to wildlife is limited. Using video recordings at a custom-built feeder, we extract key frames and evaluate the re-identification performance of our pipeline. Our unsupervised methodology combines object detection using YOLO and Grounding DINO, optical flow blur detection, image encoding with DINOv2, and clustering methods to identify representative key frames. The results indicate that our proposed key frame selection methods yield image collections which achieve high accuracy in kākā re-identification, providing a foundation for future research using media collected in more diverse and challenging environments. Through the use of artificial intelligence and computer vision, our non-invasive and efficient approach provides a valuable alternative to traditional physical tagging methods for recognising kākā individuals and therefore improving the monitoring of populations. This research contributes to developing fresh approaches in wildlife monitoring, with applications in ecology and conservation biology.

While counterfactual fairness of point predictors is well studied, its extension to prediction sets--central to fair decision-making under uncertainty--remains underexplored. On the other hand, conformal prediction (CP) provides efficient, distribution-free, finite-sample valid prediction sets, yet does not ensure counterfactual fairness. We close this gap by developing Counterfactually Fair Conformal Prediction (CF-CP) that produces counterfactually fair prediction sets. Through symmetrization of conformity scores across protected-attribute interventions, we prove that CF-CP results in counterfactually fair prediction sets while maintaining the marginal coverage property. Furthermore, we empirically demonstrate that on both synthetic and real datasets, across regression and classification tasks, CF-CP achieves the desired counterfactual fairness and meets the target coverage rate with minimal increase in prediction set size. CF-CP offers a simple, training-free route to counterfactually fair uncertainty quantification.

Benchmark scores for Large Language Models (LLMs) can be inflated by memorization of test items or near duplicates. We present a simple, protocol that probes generalization by re-evaluating models on paraphrased versions of benchmark questions. Using Mistral-7B-Instruct and Qwen2.5-7B-Instruct, we measure the accuracy gap between original and paraphrased items on ARC-Easy and ARC-Challenge. Our pipeline controls decoding, enforces multiple-choice output format, and includes a robust paraphrase-cleaning step to preserve semantics. We find that paraphrasing induces a non-trivial accuracy drop (original vs. paraphrased), consistent with prior concerns about contamination and brittle surface-form shortcuts.