Wentao Hu

Generating executable structured queries for OpenStreetMap (OSM) from natural language remains a challenging task due to the rigid syntax of OverpassQL and the steep learning curve it presents to end-users. While steering Large Language Models (LLMs) via demonstration contexts offers a promising solution, existing retrieval-based approaches typically neglect the critical constraint of token budgets. This oversight often leads to prohibitive inference costs and ignores the combinatorial nature of context selection, resulting in suboptimal performance. Furthermore, the lack of reliable benchmarks hinders rigorous evaluation in this domain. To address these challenges, we first introduce OsmNL , a robust benchmark curated via a human-in-the-loop pipeline to resolve alignment noise in prior datasets. Building upon this foundation, we formulate the construction of optimal demonstration contexts under constraints as a Budgeted Example Subset Mining ( BESM ) problem. We analyze BESM as an NP-hard optimization task that admits a submodular objective, theoretically balancing semantic relevance with structural diversity. Leveraging this property, we propose TACO ( T oken- A ware C ontext O ptimization), a principled framework that instantiates a family of budget-aware algorithms to navigate the trade-off between approximation guarantees and computational efficiency. Extensive experiments on OsmNL show that TACO establishes a new state-of-the-art, achieving superior accuracy compared to strong baselines while reducing token consumption by over 50%, thereby offering a robust and cost-efficient solution for large-scale geospatial query interfaces.

Foundation-model agents now use reusable skills for tool use, long-horizon planning, and adaptation across related tasks. The term, however, is used loosely. It may describe a prompt package, an executable workflow, a learned routine, or an artifact distributed through a repository. That looseness makes it hard to compare methods, measure progress, or discuss security and governance with precision.We study agent skills as reusable and adaptive units of competence between model capability and situated task execution. The survey separates skills from nearby constructs such as prompts, tools, memory, and policies, then organizes the literature around representation, lifecycle and orchestration, evaluation, security and governance, and application domains. Across these areas, skill quality is only one part of the story. Useful skills also depend on abstraction choices, retrieval and composition mechanisms, ecosystem structure, and infrastructure security. We treat agent skills as a research object in their own right and identify open problems in automatic induction, cross-environment transfer, longitudinal evaluation, and trustworthy sharing in open agent ecosystems. A public paper list is available at https://github.com/JinhaoShen/awesome-agent-skill-papers.

Recent advancements in diffusion models have significantly impacted content creation, leading to the emergence of personalized content synthesis (PCS). By utilizing a small set of user-provided examples featuring the same subject, PCS aims to tailor this subject to specific user-defined prompts. Over the past two years, more than 150 methods have been introduced in this area. However, existing surveys primarily focus on text-to-image generation, with few providing up-to-date summaries on PCS. This paper provides a comprehensive survey of PCS, introducing the general frameworks of PCS research, which can be categorized into test-time fine-tuning (TTF) and pre-trained adaptation (PTA) approaches. We analyze the strengths, limitations and key techniques of these methodologies. Additionally, we explore specialized tasks within the field, such as object, face and style personalization, while highlighting their unique challenges and innovations. Despite the promising progress, we also discuss ongoing challenges, including overfitting and the trade-off between subject fidelity and text alignment. Through this detailed overview and analysis, we propose future directions to further the development of PCS.

Retrieval-Augmented Generation (RAG) enhances factual accuracy by integrating external knowledge, yet it introduces a critical issue: erroneous or biased retrieval can mislead generation, compounding hallucinations, a phenomenon we term Hallucination on Hallucination. To address this, we propose Debate-Augmented RAG (DRAG), a training-free framework that integrates Multi-Agent Debate (MAD) mechanisms into both retrieval and generation stages. In retrieval, DRAG employs structured debates among proponents, opponents, and judges to refine retrieval quality and ensure factual reliability. In generation, DRAG introduces asymmetric information roles and adversarial debates, enhancing reasoning robustness and mitigating factual inconsistencies. Evaluations across multiple tasks demonstrate that DRAG improves retrieval reliability, reduces RAG-induced hallucinations, and significantly enhances overall factual accuracy. Our code is available at https://github.com/Huenao/Debate-Augmented-RAG.

Wentao Hu, Wengyu Zhang, Yiyang Jiang, Chen Jason Zhang, Xiaoyong Wei, Li Qing. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.