Automatic prompt generation plays a crucial role in enabling general-purpose multi-agent systems to perform diverse tasks autonomously. Existing methods typically evaluate prompts based on their immediate task performance, overlooking the intrinsic qualities that determine their reliability. This outcome-centric view n...
Ke Chen, Xucheng Yu, Yufei Zhou et al.· 0 citations
The proposed HyGA sheds light on a more effective, efficient, and stable attention mechanism that contains three types of gating strategies and introduces low-rank matrix decomposition and learnable attention sink to further enhance training efficiency and stability.
Zekun Zhou, Ruo-Bing Xie, Lan-Rui Wang et al.· 1 citation
Executing long-horizon tool invocations in real-world environments is severely challenged by dynamic user intent noise. Existing methods attempt robustness via implicit history scanning or text compression, yet predominantly assume perfect instructions in simplistic scenarios. Inevitably, under fluctuating contexts, ob...
Dingwei Zhu, Jiahan Li, Chengjun Pan et al.· 0 citations
DeepRubric is introduced, a data construction framework that first determines what an evidence-backed report should be evaluated on and then synthesizes aligned query--rubric pairs from those evaluation targets, achieving comparable performance to prior open state-of-the-art deep research models across three benchmarks...
Ming-Hang Zhu, Chu-Yang Wei, Jun Xu et al.· arXiv.org· 5 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Effective peer review requires adaptive investigation that responds to evidence emerging from each paper. Existing LLM-based reviewers largely organize the review process around predefined workflows, leaving limited scope for emerging evidence to reshape investigation. We propose ProReviewer, an evidence-driven review...
This work introduces MASDR-RAG (Metadata-Scoped Domain Retrieval for RAG) as a controlled testbed and finds hard partitioning to be unsafe, whereas training-free soft scoping is safe when the scope structure is real.
N. Subedi, A. Abdelaty, S. Sheshappanavar· 0 citations
Given the increased use of large language models (LLMs) for reading and seeking information and the varying needs and backgrounds of users, direct manipulation of text can be a valuable mechanism for end-user control. However, enabling direct manipulation requires that models generate text consistent with user-requeste...
Speech translation (ST) is increasingly adopted in user applications, yet its evaluation largely focuses on decontextualized testbeds and holistic quality, rather than end users' communication needs. We introduce Ouvia, an evaluation framework for measuring user-perceived usability of speech translation outputs in real...
Giuseppe Attanasio, Beatrice Savoldi, Daniel Chechelnitsky et al.· 0 citations
LLMs can plan by generating schedules directly as a Planner, or by translating a task into a formal specification for a solver as a Formalizer. Most real-world tasks are asynchronous, with non-uniform durations, concurrency, and execution-time changes, yet existing benchmarks hardly cover them. We unify asynchronous pl...
Jiayi Zhang, Jianing Yin, Ben Zhou et al.· 0 citations
Retrieval-augmented generation (RAG) has emerged as a promising paradigm for enhancing large language models (LLMs) on multi-hop question answering (QA), which requires reasoning over evidence from multiple documents. However, current multi-hop RAG approaches still struggle with evidence selection and information trans...
Yikai Zhu, Kunfeng Chen, Qihuang Zhong et al.· 0 citations
Automatic evaluation of long-form literary writing remains challenging, particularly for creativity-related dimensions such as originality and flexibility. We address the lack of large-scale supervision for structured Torrance Test of Creative Writing (TTCW)-based evaluation by constructing a dataset of 263{,}911 long-...
Jinlong Liu, Mohammed Bahja, Mark Lee· 0 citations
This work conducts a scoping review spanning four decades and develops a taxonomy of 22 recurring concerns, finding that some concerns have become less visible in literature and many lack coverage in institutional checklists.
Ruchira Dhar, Qi-Wei Peng, Anders Søgaard· 0 citations
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
