Providing high-quality feedback on writing is cognitively demanding, requiring reviewers to identify issues, suggest fixes, and ensure consistency. We introduce AnnotateGPT, a system that uses pen-based annotations as an input modality for AI agents to assist with essay feedback. AnnotateGPT enhances feedback by interpreting handwritten annotations and extending them throughout the document. One AI agent classifies the purpose of each annotation, which is confirmed or corrected by the user. A second AI agent uses the confirmed purpose to generate contextually relevant feedback for other parts of the essay. In a study with 12 novice teachers annotating essays, we compared AnnotateGPT with a baseline pen-based tool without AI support. Our findings demonstrate how reviewers used annotations to regulate AI feedback generation, refine AI suggestions, and incorporate AI-generated feedback into their review process. We highlight design implications for AI-augmented feedback systems, including balanced human-AI collaboration and using pen annotations as subtle interaction.

Watching subtitled videos in a foreign language demands sustained visual attention, which can put viewers at risk of missing content due to distraction, such as checking notifications. In this work, we introduced a gaze-aware video player that adapts playback to support attention recovery. We evaluated three gaze-aware techniques: adaptive pausing, stacked subtitles, and audio language switching (dubbing). In a comparative study with 24 participants, we evaluated these techniques against a standard video player with subtitles. While adaptive pausing improved task performance and reduced distractions, stacked subtitles helped recover reading but occasionally slowed faster readers. The benefit of dubbing was limited, resulting in additional cognitive load during the process. Ultimately, all gaze-aware interventions outperformed the standard video player. This work highlights gaze-adaptive systems that seamlessly support attention recovery into everyday viewing experiences.

Effective comprehension is essential for learning and understanding new material. However, human-generated questions often fail to cater to individual learners’ needs and interests. We propose a novel approach that leverages a gaze-driven interest model and a Large Language Model (LLM) to generate personalized comprehension questions automatically for short (∼ 10 min) educational video content. Our interest model scores each word in a subtitle. The top-scoring words are then used to generate questions using an LLM. Additionally, our system provides marginal help by offering phrase definitions (glosses) in subtitles, further facilitating learning. These methods are integrated into a prototype system, GazeQ-GPT, automatically focusing learning material on specific content that interests or challenges them, promoting more personalized learning. A user study (N = 40) shows that GazeQ-GPT prioritizes words in the fixated gloss and rewatched subtitles with higher ratings toward glossed videos. Compared to ChatGPT, GazeQ-GPT achieves higher question diversity while maintaining quality, indicating its potential to improve personalized learning experiences through dynamic content adaptation.

The growing dimensions of smartphones have intensified the challenges associated with screen reachability. Back-of-device (BoD) interaction expands the range of reachability and offers a promising solution to mitigate screen occlusion while enhancing one-handed interactions. However, much of the existing research relies on incorporating additional hardware components. In this paper, we present SwipeSense a technique for exploring the feasibility of directional swipe interactions on the back of devices, utilizing built-in inertial measurement unit (IMU) sensors and machine learning models. We conducted a user study with 12 participants who performed 9600 BoD swipes in 8 distinct directions while holding the device naturally. The results of our machine learning models indicate that various directional swipes on the back of the device can be accurately distinguished using only the built-in IMU sensors of the phone, achieving a range of model accuracy between 72% and 95%. Furthermore, we showcase potential applications for these gestures.

Many visualizations, including word clouds, cartographic labels, and word trees, encode data within the sizes of fonts. While font size can be an intuitive dimension for the viewer, using it as an encoding can introduce factors that may bias the perception of the underlying values. Viewers might conflate the size of a word's font with a word's length, the number of letters it contains, or with the larger or smaller heights of particular characters ('o' versus 'p' versus 'b'). We present a collection of empirical studies showing that such factors-which are irrelevant to the encoded values-can indeed influence comparative judgements of font size, though less than conventional wisdom might suggest. We highlight the largest potential biases, and describe a strategy to mitigate them.