Accurate canal network mapping is essential for water management, including irrigation planning and infrastructure maintenance. State-of-the-art semantic segmentation models for infrastructure mapping, such as roads, rely on large, well-annotated remote sensing datasets. However, incomplete or inadequate ground truth can hinder these learning approaches. Many infrastructure networks have graph-level properties such as reachability to a source (like canals) or connectivity (roads) that can be leveraged to improve these existing ground truth. This paper develops a novel iterative framework IGraSS, combining a semantic segmentation module-incorporating RGB and additional modalities (NDWI, DEM)-with a graph-based ground-truth refinement module. The segmentation module processes satellite imagery patches, while the refinement module operates on the entire data viewing the infrastructure network as a graph. Experiments show that IGraSS reduces unreachable canal segments from around 18% to 3%, and training with refined ground truth significantly improves canal identification. IGraSS serves as a robust framework for both refining noisy ground truth and mapping canal networks from remote sensing imagery. We also demonstrate the effectiveness and generalizability of IGraSS using road networks as an example, applying a different graph-theoretic constraint to complete road networks.

Refusal discovery is the task of identifying the full set of topics that a language model refuses to discuss. We introduce this new problem setting and develop a refusal discovery method, Iterated Prefill Crawler (IPC), that uses token prefilling to find forbidden topics. We benchmark IPC on Tulu-3-8B, an open-source model with public safety tuning data. Our crawler manages to retrieve 31 out of 36 topics within a budget of 1000 prompts. Next, we scale the crawler to a frontier model using the prefilling option of Claude-Haiku. Finally, we crawl three widely used open-weight models: Llama-3.3-70B and two of its variants finetuned for reasoning: DeepSeek-R1-70B and Perplexity-R1-1776-70B. DeepSeek-R1-70B reveals patterns consistent with censorship tuning: The model exhibits "thought suppression" behavior that indicates memorization of CCP-aligned responses. Although Perplexity-R1-1776-70B is robust to censorship, IPC elicits CCP-aligned refusals answers in the quantized model. Our findings highlight the critical need for refusal discovery methods to detect biases, boundaries, and alignment failures of AI systems.

While large language models (LLMs) can solve PhD-level reasoning problems over long context inputs, they still struggle with a seemingly simpler task: following explicit length instructions-e.g., write a 10,000-word novel. Additionally, models often generate far too short outputs, terminate prematurely, or even refuse the request. Existing benchmarks focus primarily on evaluating generations quality, but often overlook whether the generations meet length constraints. To this end, we introduce Length Instruction Following Evaluation Benchmark (LIFEBench) to comprehensively evaluate LLMs' ability to follow length instructions across diverse tasks and a wide range of specified lengths. LIFEBench consists of 10,800 instances across 4 task categories in both English and Chinese, covering length constraints ranging from 16 to 8192 words. We evaluate 26 widely-used LLMs and find that most models reasonably follow short-length instructions but deteriorate sharply beyond a certain threshold. Surprisingly, almost all models fail to reach the vendor-claimed maximum output lengths in practice, as further confirmed by our evaluations extending up to 32K words. Even long-context LLMs, despite their extended input-output windows, counterintuitively fail to improve length-instructions following. Notably, Reasoning LLMs outperform even specialized long-text generation models, achieving state-of-the-art length following. Overall, LIFEBench uncovers fundamental limitations in current LLMs' length instructions following ability, offering critical insights for future progress.

Autonomous navigation in off-road environments remains a significant challenge in field robotics, particularly for Unmanned Ground Vehicles (UGVs) tasked with search and rescue, exploration, and surveillance. Effective long-range planning relies on the integration of onboard perception systems with prior environmental knowledge, such as satellite imagery and LiDAR data. This work introduces Trailblazer, a novel framework that automates the conversion of multi-modal sensor data into costmaps, enabling efficient path planning without manual tuning. Unlike traditional approaches, Trailblazer leverages imitation learning and a differentiable A* planner to learn costmaps directly from expert demonstrations, enhancing adaptability across diverse terrains. The proposed methodology was validated through extensive real-world testing, achieving robust performance in dynamic and complex environments, demonstrating Trailblazer's potential for scalable, efficient autonomous navigation.

Deliberative democracy depends on carefully designed institutional frameworks -- such as participant selection, facilitation methods, and decision - making mechanisms -- that shape how deliberation performs . However, identifying optimal institutional designs for specific contexts remains challenging when relying solely on real - world observations or laboratory experiments: they can be expensive, ethically and methodologically tricky, or too limited in scale to give us clear answers . Computational experiments offer a complementary approach, enabling researchers to conduct large - scale investigations while systematically analyzing complex dynamics, emergent and unexpected collective behavior, and risks or opportunities associated with novel democratic designs . Therefore, this paper explores Digital Twin (DT) technology as a computational testing ground for deliberative systems (with potential applicability to broader institutional analysis) . By constructing dynamic models that simulate real - world deliberation, DTs allow researchers and policymakers to rigorously test "what - if" scenarios across diverse institutional configurations in a controlled virtual environment. This approach facilitates evidence - based assessment of novel designs using synthetically generated data, bypassing the constraints of real - world or lab - based experimentation, and without societal disruption. The paper also discusses the limitations of this new methodological approach and suggest s where future research should focus .

Responsible AI (rAI) guidance increasingly promotes stakeholder involvement (SHI) during AI development. At the same time, SHI is already common in commercial software development, but with potentially different foci. This study clarifies the extent to which established SHI practices are able to contribute to rAI efforts as well as potential disconnects -- essential insights to inform and tailor future interventions that further shift industry practice towards rAI efforts. First, we analysed 56 rAI guidance documents to identify why SHI is recommended (i.e. its expected benefits for rAI) and uncovered goals such as redistributing power, improving socio-technical understandings, anticipating risks, and enhancing public oversight. To understand why and how SHI is currently practised in commercial settings, we then conducted an online survey (n=130) and semi-structured interviews (n=10) with AI practitioners. Our findings reveal that SHI in practice is primarily driven by commercial priorities (e.g. customer value, compliance) and several factors currently discourage more rAI-aligned SHI practices. This suggests that established SHI practices are largely not contributing to rAI efforts. To address this disconnect, we propose interventions and research opportunities to advance rAI development in practice.

Large Language Models (LLMs) have shown strong inductive reasoning ability across various domains, but their reliability is hindered by the outdated knowledge and hallucinations. Retrieval-Augmented Generation mitigates these issues by grounding LLMs with external knowledge; however, most existing RAG pipelines rely on unstructured text, limiting interpretability and structured reasoning. Knowledge graphs, which represent facts as relational triples, offer a more structured and compact alternative. Recent studies have explored integrating knowledge graphs with LLMs for knowledge graph question answering (KGQA), with a significant proportion adopting the retrieve-then-reasoning paradigm. In this framework, graph-based retrievers have demonstrated strong empirical performance, yet they still face challenges in generalization ability. In this work, we propose RAPL, a novel framework for efficient and effective graph retrieval in KGQA. RAPL addresses these limitations through three aspects: (1) a two-stage labeling strategy that combines heuristic signals with parametric models to provide causally grounded supervision; (2) a model-agnostic graph transformation approach to capture both intra- and inter-triple interactions, thereby enhancing representational capacity; and (3) a path-based reasoning strategy that facilitates learning from the injected rational knowledge, and supports downstream reasoner through structured inputs. Empirically, RAPL outperforms state-of-the-art methods by $2.66\%-20.34\%$, and significantly reduces the performance gap between smaller and more powerful LLM-based reasoners, as well as the gap under cross-dataset settings, highlighting its superior retrieval capability and generalizability. Codes are available at: https://github.com/tianyao-aka/RAPL.

Clinical trials are the bedrock of medical practice. They provide a scientifically rigorous way to test the safety and efficacy of new treatments, drugs, and medical devices. Considering the substantial investment required for clinical trials, with Phase III trials often exceeding $500 million and lasting several years [Sertkaya et al., 2024], it is crucial to conduct them with utmost efficiency. Furthermore, they represent a cornerstone in the pharmaceutical industry's ongoing commitment to enhancing patient care strategies. To inform evidence-based practice, trials should accurately represent the diversity and complexity of real-world patient populations, thereby strengthening the evidence for new treatments.

A concentrated, nine-minute-long Russian drone attack on Ukraine's second largest city of Kharkiv in the middle of the night killed six people and injured 64, including nine children, Ukrainian officials said on Wednesday. The overnight attack followed Russia's two biggest air assaults of the war on Ukraine this week, part of intensified bombardments that Moscow says are retaliatory measures for Kyiv's recent attacks in Russia. Elsewhere, two southern Ukrainian regions, Mykolaiv and Kherson, were left without electricity on Wednesday after Russian forces attacked an energy facility, the governors said.

Fox News' Alex Hogan reports on one of the largest Russian attacks on Ukraine since the war began. Fox News contributor Mike Pompeo also breaks down the Trump administration's travel ban and discusses the U.S. role in potential peace talks. North Korean ballistic missiles once again rained down over Ukraine this week as the war with Russia continues to rage, prompting President Volodymyr Zelenskyy to renew warnings that the threat posed by the Moscow-Pyongyang alliance is "extremely dangerous" for Europe and Asia alike. "The longer this war continues on our territory, the more warfare technologies evolve, and the greater the threat will be to everyone," Zelenskyy said Tuesday. "This must be addressed now, not when thousands of upgraded Shahed drones and ballistic missiles begin to threaten Seoul and Tokyo." Zelenskyy's warning came just one day after Ukraine's military intelligence chief, Kyrylo Budanov, confirmed in an interview with The War Zone that Russia has significantly improved North Korea's KN-23 ballistic missiles.