Related papers: Human-Centric Evaluation for Foundation Models
The rapid advancement of large language models (LLMs) demands increasingly reliable evaluation, yet current centralized evaluation suffers from opacity, overfitting, and hardware-induced variance. Our empirical analysis reveals an alarming…
To facilitate the widespread acceptance of AI systems guiding decision-making in real-world applications, it is key that solutions comprise trustworthy, integrated human-AI systems. Not only in safety-critical applications such as…
Test-time scaling has significantly improved how AI models solve problems, yet current methods often get stuck in repetitive, incorrect patterns of thought. We introduce HEART, a framework that uses emotional cues to guide the model's…
Search engines, as cognitive partners, reshape how individuals evaluate their cognitive abilities. This study examines how search tool access influences cognitive self-esteem (CSE)-users' self-perception of cognitive abilities -- through…
Large language models (LLMs) such as GPT-5 and Gemini 3 have pushed the frontier of automated reasoning and code generation. Yet current benchmarks emphasize accuracy and output quality, neglecting a critical dimension: efficiency of token…
Scientific discovery is an inherently creative and uncertain process, requiring reasoning beyond the recall of known knowledge. While many benchmarks have been proposed to evaluate large language model (LLM) performance on deep research…
Foundation models excel at single-turn reasoning but struggle with multi-turn exploration in dynamic environments, a requirement for many real-world challenges. We evaluated these models on their ability to learn from experience, adapt, and…
Evaluating the reasoning abilities of large language models (LLMs) solely from final answers can obscure failures in intermediate steps, especially in multi-hop QA benchmarks without step-level annotations. To address this gap, we introduce…
We review human evaluation practices in automatic, speech-driven 3D gesture generation and find a lack of standardisation and frequent use of flawed experimental setups. This leads to a situation where it is impossible to know how different…
The advent of foundation models (FMs) as an emerging suite of AI techniques has struck a wave of opportunities in computational healthcare. The interactive nature of these models, guided by pre-training data and human instructions, has…
Code generation is a core capability of large language models (LLMs), yet mainstream benchmarks (e.g., APPs and LiveCodeBench) contain questions with medium-level difficulty and pose no challenge to advanced LLMs. To better reflected the…
Failure to consider the characteristics, limitations, and abilities of diverse end-users during mobile apps development may lead to problems for end-users such as accessibility and usability issues. We refer to this class of problems as…
Deep research -- producing comprehensive, citation-grounded reports by searching and synthesizing information from hundreds of live web sources -- marks an important frontier for agentic systems. To rigorously evaluate this ability, four…
With the surge in attention to Egocentric Hand-Object Interaction (Ego-HOI), large-scale datasets such as Ego4D and EPIC-KITCHENS have been proposed. However, most current research is built on resources derived from third-person video…
Language Models (LMs) continue to advance, improving response quality and coherence. Given Internet-scale training datasets, LMs have likely encountered much of what users may ask them to generate in some form during their training. A…
Evaluating concept customization is challenging, as it requires a comprehensive assessment of fidelity to generative prompts and concept images. Moreover, evaluating multiple concepts is considerably more difficult than evaluating a single…
User preferences evolve across months of interaction, and tracking them requires inferring when a stated preference has been changed by a subsequent life event. We define this problem as long-horizon personalization and observe that…
As automated vehicles (AVs) increasingly integrate into mixed-traffic environments, evaluating their interaction with human-driven vehicles (HDVs) becomes critical. In most research focused on developing new AV control algorithms…
Humans consistently outperform state-of-the-art AI models in action recognition, particularly in challenging real-world conditions involving low resolution, occlusion, and visual clutter. Understanding the sources of this performance gap is…
As Artificial Intelligence (AI) becomes integral to software development, understanding the social and cooperative dynamics that affect AI-driven organizational change is important. Yet, despite AI's rapid progress and influence, the human…