Solving long-horizon goal-conditioned tasks remains a significant challenge in reinforcement learning (RL). Hierarchical reinforcement learning (HRL) addresses this by decomposing tasks into more manageable sub-tasks, but the automatic discovery of the hierarchy and the joint training of multi-level policies often suffer from instability and can lack theoretical guarantees. In this paper, we introduce Reinforcement Learning with Anticipation (RLA), a principled and potentially scalable framework designed to address these limitations. The RLA agent learns two synergistic models: a low-level, goal-conditioned policy that learns to reach specified subgoals, and a high-level anticipation model that functions as a planner, proposing intermediate subgoals on the optimal path to a final goal. The key feature of RLA is the training of the anticipation model, which is guided by a principle of value geometric consistency, regularized to prevent degenerate solutions. We present proofs that RLA approaches the globally optimal policy under various conditions, establishing a principled and convergent method for hierarchical planning and execution in long-horizon goal-conditioned tasks.

Educational Knowledge Graphs (EduKGs) organize various learning entities and their relationships to support structured and adaptive learning. Prerequisite relationships (PRs) are critical in EduKGs for defining the logical order in which concepts should be learned. However, the current EduKG in the MOOC platform CourseMapper lacks explicit PR links, and manually annotating them is time-consuming and inconsistent. To address this, we propose an unsupervised method for automatically inferring concept PRs without relying on labeled data. We define ten criteria based on document-based, Wikipedia hyperlink-based, graph-based, and text-based features, and combine them using a voting algorithm to robustly capture PRs in educational content. Experiments on benchmark datasets show that our approach achieves higher precision than existing methods while maintaining scalability and adaptability, thus providing reliable support for sequence-aware learning in CourseMapper.

The automatic construction of Educational Knowledge Graphs (EduKGs) is essential for domain knowledge modeling by extracting meaningful representations from learning materials. Despite growing interest, identifying a scalable and reliable approach for automatic EduKG generation remains a challenge. In an attempt to develop a unified and robust pipeline for automatic EduKG construction, in this study we propose a pipeline for automatic EduKG construction from PDF learning materials. The process begins with generating slide-level EduKGs from individual pages/slides, which are then merged to form a comprehensive EduKG representing the entire learning material. We evaluate the accuracy of the EduKG generated from the proposed pipeline in our MOOC platform, CourseMapper. The observed accuracy, while indicative of partial success, is relatively low particularly in the educational context, where the reliability of knowledge representations is critical for supporting meaningful learning. To address this, we introduce targeted optimizations across multiple pipeline components. The optimized pipeline achieves a 17.5% improvement in accuracy and a tenfold increase in processing efficiency. Our approach offers a holistic, scalable and end-to-end pipeline for automatic EduKG construction, adaptable to diverse educational contexts, and supports improved semantic representation of learning content.

Abstract: The use of large language models (LLMs) in bioethical, scientific, and medical writing remains controversial. While there is broad agreement in some circles that LLMs cannot count as authors, there is no consensus about whether and how humans using LLMs can count as authors. In many fields, authorship is distributed among large teams of researchers, some of whom -- including paradigmatic "senior authors" who guide and determine the scope of a project and ultimately vouch for its integrity -- may not write a singl e word. In this paper, we argue that LLM use (under specific conditions) is analogous to a form of senior authorship. On this view, the use of LLMs, even to generate complete drafts of research papers, can be considered a legitimate form of authorship according to the accepted criteria in many fields. We conclude that either such use should be recognized as legitimate, or current criteria for authorship require fundamental revision. AI use declaration: Chat GPT version 5 was used to help format Box 1. AI wa s not used for any other part of the preparation or writing of this manuscript. This is a pre print of a paper that has been submitted to a journal. It has not yet gone through peer review. Authorship Without Writing: Large Language Models and the "Senior Author" Analogy Clint Hurshman, Sebastian Porsdam Mann, Julian Savulescu, Brian D. Earp I. Introduction The use of large language models (LLMs) in bioethics as well as scientific and medical writing continues to be controversial. Thus far, there has been broad agreement -- for example, among medical publishers -- that LLMs cannot count as authors, but there is still no consensus about the status of LLM - assisted text production as a form of writing, and by extension, the status of LLM users as authors. Here, we contribute to this debate by exploring -- and drawing analogies to -- the collaborative nature of writing, and t he distributed character of authorship, in many domains of research.

Automatic emotion recognition has become increasingly important with the rise of AI, especially in fields like healthcare, education, and automotive systems. However, there is a lack of multimodal datasets, particularly involving body motion and physiological signals, which limits progress in the field. To address this, the MVRS dataset is introduced, featuring synchronized recordings from 13 participants aged 12 to 60 exposed to VR based emotional stimuli (relaxation, fear, stress, sadness, joy). Data were collected using eye tracking (via webcam in a VR headset), body motion (Kinect v2), and EMG and GSR signals (Arduino UNO), all timestamp aligned. Participants followed a unified protocol with consent and questionnaires. Features from each modality were extracted, fused using early and late fusion techniques, and evaluated with classifiers to confirm the datasets quality and emotion separability, making MVRS a valuable contribution to multimodal affective computing.

The advancement of Object Detection (OD) using Deep Learning (DL) is often hindered by the significant challenge of acquiring large, accurately labeled datasets, a process that is time-consuming and expensive. While techniques like Active Learning (AL) can reduce annotation effort by intelligently querying informative samples, they often lack transparency, limit the strategic insight of human experts, and may overlook informative samples not aligned with an employed query strategy. To mitigate these issues, Human-in-the-Loop (HITL) approaches integrating human intelligence and intuition throughout the machine learning life-cycle have gained traction. Leveraging Visual Analytics (VA), effective interfaces can be created to facilitate this human-AI collaboration. This thesis explores the intersection of these fields by developing and investigating "VILOD: A Visual Interactive Labeling tool for Object Detection". VILOD utilizes components such as a t-SNE projection of image features, together with uncertainty heatmaps and model state views. Enabling users to explore data, interpret model states, AL suggestions, and implement diverse sample selection strategies within an iterative HITL workflow for OD. An empirical investigation using comparative use cases demonstrated how VILOD, through its interactive visualizations, facilitates the implementation of distinct labeling strategies by making the model's state and dataset characteristics more interpretable (RQ1). The study showed that different visually-guided labeling strategies employed within VILOD result in competitive OD performance trajectories compared to an automated uncertainty sampling AL baseline (RQ2). This work contributes a novel tool and empirical insight into making the HITL-AL workflow for OD annotation more transparent, manageable, and potentially more effective.

Large reasoning models (LRMs) have exhibited strong performance on complex reasoning tasks, with further gains achievable through increased computational budgets at inference. However, current test-time scaling methods predominantly rely on redundant sampling, ignoring the historical experience utilization, thereby limiting computational efficiency. To overcome this limitation, we propose Sticker-TTS, a novel test-time scaling framework that coordinates three collaborative LRMs to iteratively explore and refine solutions guided by historical attempts. At the core of our framework are distilled key conditions-termed stickers-which drive the extraction, refinement, and reuse of critical information across multiple rounds of reasoning. To further enhance the efficiency and performance of our framework, we introduce a two-stage optimization strategy that combines imitation learning with self-improvement, enabling progressive refinement. Extensive evaluations on three challenging mathematical reasoning benchmarks, including AIME-24, AIME-25, and OlymMATH, demonstrate that Sticker-TTS consistently surpasses strong baselines, including self-consistency and advanced reinforcement learning approaches, under comparable inference budgets. These results highlight the effectiveness of sticker-guided historical experience utilization. Our code and data are available at https://github.com/RUCAIBox/Sticker-TTS.

We introduce perioperation, a paradigm for robotic data collection that sensorizes and records human manipulation while maximizing the transferability of the data to real robots. We implement this paradigm in DEXOP, a passive hand exoskeleton designed to maximize human ability to collect rich sensory (vision + tactile) data for diverse dexterous manipulation tasks in natural environments. DEXOP mechanically connects human fingers to robot fingers, providing users with direct contact feedback (via proprioception) and mirrors the human hand pose to the passive robot hand to maximize the transfer of demonstrated skills to the robot. The force feedback and pose mirroring make task demonstrations more natural for humans compared to teleoperation, increasing both speed and accuracy. We evaluate DEXOP across a range of dexterous, contact-rich tasks, demonstrating its ability to collect high-quality demonstration data at scale. Policies learned with DEXOP data significantly improve task performance per unit time of data collection compared to teleoperation, making DEXOP a powerful tool for advancing robot dexterity. Our project page is at https://dex-op.github.io.

Quadruped-based mobile manipulation presents significant challenges in robotics due to the diversity of required skills, the extended task horizon, and partial observability. After presenting a multi-stage pick-and-place task as a succinct yet sufficiently rich setup that captures key desiderata for quadruped-based mobile manipulation, we propose an approach that can train a visuo-motor policy entirely in simulation, and achieve nearly 80\% success in the real world. The policy efficiently performs search, approach, grasp, transport, and drop into actions, with emerged behaviors such as re-grasping and task chaining. We conduct an extensive set of real-world experiments with ablation studies highlighting key techniques for efficient training and effective sim-to-real transfer. Additional experiments demonstrate deployment across a variety of indoor and outdoor environments. Demo videos and additional resources are available on the project page: https://horizonrobotics.github.io/gail/SLIM.

The impressive capabilities of Large Language Models (LLMs) have fueled the notion that synthetic agents can serve as substitutes for real participants in human-subject research. In an effort to evaluate the merits of this claim, social science researchers have largely focused on whether LLM-generated survey data corresponds to that of a human counterpart whom the LLM is prompted to represent. In contrast, we address a more fundamental question: Do agents maintain internal consistency, retaining similar behaviors when examined under different experimental settings? To this end, we develop a study designed to (a) reveal the agent's internal state and (b) examine agent behavior in a basic dialogue setting. This design enables us to explore a set of behavioral hypotheses to assess whether an agent's conversation behavior is consistent with what we would expect from their revealed internal state. Our findings on these hypotheses show significant internal inconsistencies in LLMs across model families and at differing model sizes. Most importantly, we find that, although agents may generate responses matching those of their human counterparts, they fail to be internally consistent, representing a critical gap in their capabilities to accurately substitute for real participants in human-subject research. Our simulation code and data are publicly accessible.