Why we think this paper is great for you:
This paper directly explores the implementation and evaluation of an AI assistant with human involvement, offering practical insights into building effective interactive AI systems. You will find its focus on real-world application particularly relevant for understanding human-AI collaboration.

Abstract
Generative AI assistants offer significant potential to enhance productivity, streamline information access, and improve user experience in enterprise contexts. In this work, we present Summit Concierge, a domain-specific AI assistant developed for Adobe Summit. The assistant handles a wide range of event-related queries and operates under real-world constraints such as data sparsity, quality assurance, and rapid deployment. To address these challenges, we adopt a human-in-the-loop development workflow that combines prompt engineering, retrieval grounding, and lightweight human validation. We describe the system architecture, development process, and real-world deployment outcomes. Our experience shows that agile, feedback-driven development enables scalable and reliable AI assistants, even in cold-start scenarios.

AI Summary

• Agile, feedback-driven development, including daily human-in-the-loop reviews and continuous monitoring of user feedback, enables rapid identification and resolution of issues in production. [3]
• Product Knowledge: General information queries answered using unstructured content, such as guidebooks. [3]
• Human-in-the-loop (HITL) development is crucial for rapidly deploying reliable, domain-specific generative AI assistants, especially in cold-start scenarios with data sparsity. [2]
• Combining prompt engineering, documentation-aware retrieval, and synthetic data augmentation effectively bootstraps AI assistants when historical interaction data is limited. [2]
• A multi-faceted evaluation framework, integrating correctness scoring, side-by-side comparisons, and brand compliance checks with LLM-as-judge and human validation, ensures high-quality responses. [2]
• Reasoning-oriented LLM prompt rewriting with chain-of-thought and few-shot examples significantly improves handling of ambiguous multi-turn dialogues, reducing rewrite error rates and enhancing routing accuracy. [2]
• The system demonstrated practical benefits in a real-world deployment, improving user experience and reducing operational overhead by effectively managing event-related queries. [2]
• Summit Concierge: A domain-specific generative AI assistant developed for Adobe Summit to handle event-related queries. [2]
• Human-in-the-loop (HITL) development paradigm: An iterative development approach that integrates human expertise for data curation, response validation, and quality monitoring to ensure reliability and rapid iteration. [2]
• Leveraging tools like SQLSynth for programmatic question generation from database schemas guarantees in-scope and diverse evaluation datasets, critical for stress testing intent detection and NL2SQL modules. [1]

Why we think this paper is great for you:
This work investigates the crucial alignment between AI models and human values, which is essential for developing responsible and trustworthy AI. It provides valuable perspectives on how human judgment guides the ethical development of AI systems.

Abstract
Large Language Models (LLMs) are increasingly employed in software engineering tasks such as requirements elicitation, design, and evaluation, raising critical questions regarding their alignment with human judgments on responsible AI values. This study investigates how closely LLMs' value preferences align with those of two human groups: a US-representative sample and AI practitioners. We evaluate 23 LLMs across four tasks: (T1) selecting key responsible AI values, (T2) rating their importance in specific contexts, (T3) resolving trade-offs between competing values, and (T4) prioritizing software requirements that embody those values. The results show that LLMs generally align more closely with AI practitioners than with the US-representative sample, emphasizing fairness, privacy, transparency, safety, and accountability. However, inconsistencies appear between the values that LLMs claim to uphold (Tasks 1-3) and the way they prioritize requirements (Task 4), revealing gaps in faithfulness between stated and applied behavior. These findings highlight the practical risk of relying on LLMs in requirements engineering without human oversight and motivate the need for systematic approaches to benchmark, interpret, and monitor value alignment in AI-assisted software development.

Why we think this paper is great for you:
You'll appreciate this paper's focus on ensuring trustworthy and sustainable AI progress by understanding the capabilities and risks of autonomous AI systems. It highlights the importance of oversight in advanced AI applications.

Abstract
Understanding the current capabilities and risks of AI Scientist systems is essential for ensuring trustworthy and sustainable AI-driven scientific progress while preserving the integrity of the academic ecosystem. To this end, we develop Jr. AI Scientist, a state-of-the-art autonomous AI scientist system that mimics the core research workflow of a novice student researcher: Given the baseline paper from the human mentor, it analyzes its limitations, formulates novel hypotheses for improvement, validates them through rigorous experimentation, and writes a paper with the results. Unlike previous approaches that assume full automation or operate on small-scale code, Jr. AI Scientist follows a well-defined research workflow and leverages modern coding agents to handle complex, multi-file implementations, leading to scientifically valuable contributions. For evaluation, we conducted automated assessments using AI Reviewers, author-led evaluations, and submissions to Agents4Science, a venue dedicated to AI-driven scientific contributions. The findings demonstrate that Jr. AI Scientist generates papers receiving higher review scores than existing fully automated systems. Nevertheless, we identify important limitations from both the author evaluation and the Agents4Science reviews, indicating the potential risks of directly applying current AI Scientist systems and key challenges for future research. Finally, we comprehensively report various risks identified during development. We hope these insights will deepen understanding of current progress and risks in AI Scientist development.

Why we think this paper is great for you:
This paper delves into systems for human control of humanoid robots, offering insights into direct human interaction with complex autonomous entities. It explores the practical aspects of human-system interfaces in robotics.

Abstract
Large-scale data has driven breakthroughs in robotics, from language models to vision-language-action models in bimanual manipulation. However, humanoid robotics lacks equally effective data collection frameworks. Existing humanoid teleoperation systems either use decoupled control or depend on expensive motion capture setups. We introduce TWIST2, a portable, mocap-free humanoid teleoperation and data collection system that preserves full whole-body control while advancing scalability. Our system leverages PICO4U VR for obtaining real-time whole-body human motions, with a custom 2-DoF robot neck (cost around $250) for egocentric vision, enabling holistic human-to-humanoid control. We demonstrate long-horizon dexterous and mobile humanoid skills and we can collect 100 demonstrations in 15 minutes with an almost 100% success rate. Building on this pipeline, we propose a hierarchical visuomotor policy framework that autonomously controls the full humanoid body based on egocentric vision. Our visuomotor policy successfully demonstrates whole-body dexterous manipulation and dynamic kicking tasks. The entire system is fully reproducible and open-sourced at https://yanjieze.com/TWIST2 . Our collected dataset is also open-sourced at https://twist-data.github.io .

Why we think this paper is great for you:
This paper discusses the limitations of autonomous AI agents in scientific discovery, implicitly suggesting areas where human guidance or intervention could enhance long-term effectiveness. It provides context for where human oversight becomes critical.

Abstract
Data-driven scientific discovery requires iterative cycles of literature search, hypothesis generation, and data analysis. Substantial progress has been made towards AI agents that can automate scientific research, but all such agents remain limited in the number of actions they can take before losing coherence, thus limiting the depth of their findings. Here we present Kosmos, an AI scientist that automates data-driven discovery. Given an open-ended objective and a dataset, Kosmos runs for up to 12 hours performing cycles of parallel data analysis, literature search, and hypothesis generation before synthesizing discoveries into scientific reports. Unlike prior systems, Kosmos uses a structured world model to share information between a data analysis agent and a literature search agent. The world model enables Kosmos to coherently pursue the specified objective over 200 agent rollouts, collectively executing an average of 42,000 lines of code and reading 1,500 papers per run. Kosmos cites all statements in its reports with code or primary literature, ensuring its reasoning is traceable. Independent scientists found 79.4% of statements in Kosmos reports to be accurate, and collaborators reported that a single 20-cycle Kosmos run performed the equivalent of 6 months of their own research time on average. Furthermore, collaborators reported that the number of valuable scientific findings generated scales linearly with Kosmos cycles (tested up to 20 cycles). We highlight seven discoveries made by Kosmos that span metabolomics, materials science, neuroscience, and statistical genetics. Three discoveries independently reproduce findings from preprinted or unpublished manuscripts that were not accessed by Kosmos at runtime, while four make novel contributions to the scientific literature.

Why we think this paper is great for you:
This research focuses on evaluating the problem-solving abilities of AI agents in complex tasks, which often necessitates human-defined metrics or expert assessment. It offers insights into how human evaluation contributes to improving AI performance.

Abstract
Large language model (LLM) agents have exhibited strong problem-solving competence across domains like research and coding. Yet, it remains underexplored whether LLM agents can tackle compounding real-world problems that require a diverse set of tools to complete. Given a broad, heterogeneous tool repository, LLM agents must not only select appropriate tools based on task planning analysis but also strategically schedule the execution order to ensure efficiency. This paper introduces TPS-Bench to benchmark the ability of LLM agents in solving such problems that demand Tool Planning and Scheduling. TPS-Bench collects 200 compounding tasks of two difficulty levels, based on a tool repository containing hundreds of model context protocol (MCP) tools. In particular, each task is composed of multiple subtasks, such as web search, map navigation, calendar checking, etc., and each subtask can be completed by a basic tool. Our evaluation emphasizes both task completion rate and efficiency. The empirical studies on popular closed-source and open-source LLMs indicate that most models can perform reasonable tool planning, but differ in scheduling. For example, GLM-4.5 achieves an outperforming task completion rate of 64.72% with extensive sequential tool calls, hence suffering from significantly long execution time. By contrast, GPT-4o prioritizes parallel tool calls but achieves only a 45.08% completion rate. Considering reinforcement learning (RL) can be a viable way to improve the scheduling efficiency without compromising performance, we perform an initial study on Qwen3-1.7B and witness a 14% reduction in execution time alongside a 6% gain in task completion rate based on rarely 100 RL training samples. Our code is available https://github.com/hanwenxu1/mcp-agent.

Why we think this paper is great for you:
This paper addresses the brittleness of AI agents in complex environments, a common challenge that often requires human intervention or feedback to ensure robustness. It highlights scenarios where human input can significantly improve agent reliability.

Abstract
LLM agents excel in compact environments requiring deep reasoning but remain brittle when operating in broader, more complex contexts that demand robustness across diverse tools and schemas. Building bespoke environments for training is heavy, brittle, and limits progress. In this paper, we demonstrate that LLMs can simulate realistic environment feedback without access to actual testbed data or APIs. Inspired by this capability, we propose two frameworks: Simia-SFT, a pipeline that synthesizes SFT data by amplifying small seed sets into diverse trajectories in an environment-agnostic manner, and Simia-RL, a framework that enables RL training without real environment implementations through LLM-simulated feedback. Fine-tuning open models yields consistent improvements across multiple benchmarks, surpassing GPT-4o and approaching o4-mini on $\tau^2$-Bench. Together, Simia-SFT and Simia-RL enable scalable agent training without environment engineering, replacing heavy and brittle implementations with flexible LLM-based simulation.