Human-Computer Interaction

Ruben Laukkonen, Seb Krier, Chloé Bakalar et al.

For AI alignment researchers, this paper introduces a complementary agenda to safety-focused alignment, aiming to broaden the scope of alignment to include proactive support for human flourishing.

Matthew O. Jackson, Qiaozhu Me, Stephanie W. Wang et al.

This paper proposes a new interdisciplinary field for researchers interested in the societal impact of AI and the application of AI to social sciences.

Peter Brodeur, Jacob M. Koshy, Anil Palepu et al.

This addresses the challenge of translating AI diagnostic tools into real-world clinical practice for primary care settings, though it's an incremental step with promising initial results.

Kanzhi Cheng, Zehao Li, Zheng Ma et al.

For researchers in mobile agent automation, OpenMobile provides an open-source alternative to closed-source systems, enabling reproducible training and benchmarking.

Suhyun Lee, Palakorn Achananuparp, Neemesh Yadav et al.

For developers of LLM-based mental health tools, this provides a more diagnostic safety evaluation method, though it is an incremental improvement over existing static benchmarks.

Yue Huang, Yuchen Ma, Jiayi Ye et al.

For researchers evaluating LLMs in interactive narrative settings, this provides a more comprehensive benchmark than existing static or single-turn evaluations.

Kuangshi Ai, Haichao Miao, Kaiyuan Tang et al.

This addresses the need for reproducible evaluation in the scientific visualization community, though it is incremental as it builds on existing agentic systems.

Bo Ni, Leyao Wang, Yu Wang et al.

For researchers in conversational AI, this survey provides a structured overview of LLM-based user simulation, but it is an incremental contribution as it primarily organizes existing work without introducing new methods or results.

Yichen Feng, Yuetai Li, Chunjiang Liu et al.

For researchers and developers of multimodal AI systems, this work exposes a measurable gap between current models and expert aesthetic judgment, providing a benchmark to track progress.

Andrey Moskalenko, Alexey Bryncev, Ivan Kosmynin et al.

This challenge provides a new benchmark and dataset for the video saliency prediction community, but is an incremental contribution as it follows the format of previous NTIRE challenges.

Xiaoze Liu, Ruowang Zhang, Amir H. Abdi et al.

For developers of proactive AI agents, this work offers a practical, efficient alternative to LLM-based event processing that is deployable on-device.

Pao Siangliulue, Jonathan Bragg, Doug Downey et al.

This addresses the challenge for scientific researchers who manually query systems and struggle with exhaustive reports, offering a proactive assistant to improve efficiency, though it is incremental in enhancing existing deep research systems.

Shangqing Tu, Yanjia Li, Keyu Chen et al.

For educators lacking programming skills, MAIC-UI lowers the barrier to creating interactive courseware with pedagogical accuracy and rapid iteration, addressing a practical bottleneck in educational content creation.

Amrita Mazumdar, Seonwook Park, Rajarshi Roy et al.

For researchers developing multimodal conversational agents, this benchmark provides a systematic evaluation framework to identify critical failure modes in full-duplex audiovisual interaction.

Yanick Zengaffinen, Andreas Opedal, Donya Rooein et al.

This work addresses the challenge of creating plausible educational content for AI in education, though it is incremental as it builds on existing LLM capabilities.

Zichao Wang, Alexa Siu

This work addresses the problem of using LLMs for product discovery, showing incremental progress by validating their limits and potential in design research workflows.

Ming Zhu, Juntao Tan, Rithesh Murthy et al.

For researchers evaluating LLM-based agents, this work provides a more realistic user simulation framework that exposes critical failure modes masked by existing cooperative simulators.

AlayaWorld Team, Kaipeng Zhang, Chuanhao Li et al.

This work provides a practical foundation for researchers and developers to build and deploy long-horizon, playable video world models, addressing the high cost and inflexibility of traditional game development.

Yiming Zhao, Yu Zeng, Wenxuan Huang et al.

For researchers and practitioners in video understanding, this work addresses the bottleneck of precise spatiotemporal localization by enabling proactive visual evidence retrieval, offering a significant performance gain over existing methods.

Aakriti Kumar, Nalin Poungpeth, Diyi Yang et al.

This addresses the problem of ineffective empathic expression for individuals in personal and workplace contexts, offering a scalable AI-based intervention, though it is incremental in applying existing LLM technology to a new domain.