The fungi prefer to grow in cow dung. A newly described African species in the magic mushroom family confirms its evolutionary origin. 'Psilocybe ochraceocentrata' is found growing on cattle dung in the grasslands of southern Africa and Zimbabwe. Breakthroughs, discoveries, and DIY tips sent six days a week. The discovery of a new magic mushroom species in Africa is forcing mycologists to take another look at the famous psychedelic fungi's evolutionary history.

The vast amount of data required to model wildfires at a larger scale presents difficulties in the collection and curation of the data.

Text-to-Image (T2I) has witnessed significant advancements, demonstrating superior performance for various generative tasks.

Annotateddimensionsinclude Valence, Arousal and Emotion label, with annotations gathered using Prolific.

Could Ukraine hold a presidential election right now? Will Europe use frozen Russian assets to fund war? How can Ukraine rebuild China ties? 'Ukraine is running out of men, money and time' A Russian drone strike hit a residential building in the Ukrainian city of Dnipro, injuring seven people and causing significant damage, according to local officials. Residents said the blast shattered windows and sparked a fire in the apartment block.

President Trump's favourite word is tariffs. He reminded the world of that in his pre-Christmas address to the nation. With the world still unwrapping the tariffs gift from the first year of his second term in office, he said they were bringing jobs, higher wages and economic growth to the US. What is less debatable is that they've refashioned the global economy, and will continue to do so into 2026. The International Monetary Fund (IMF) says that although the tariff shock is smaller than originally announced, it is a key reason why it now expects the rate of global economic growth to slow to 3.1% in 2026.

Science Archaeology World's oldest poison-tipped arrow discovered in South Africa The 60,000-year-old relic contains traces of a toxic onion. Breakthroughs, discoveries, and DIY tips sent every weekday. For thousands of years, hunters around the world have employed poison-tipped arrows to assist in taking down prey. For example, the curare plant poisons used by South and Central American hunters paralyzes the respiratory system. Meanwhile, inhabitants of the Kalahari Desert have relied on the toxins harvested from beetle larvae .

The Swampy Black Iguana is the oldest specimen living at the Iguana Station scientific station, where they have a breeding and conservation project for black spiny-tailed iguanas. This species, endemic to Utila, is in danger of extinction. The Utila Iguana Conservation Project seeks to ensure the survival of this species. Breakthroughs, discoveries, and DIY tips sent every weekday. With the turning of the calendar comes a new year and new vulnerable endangered plant and animal species to keep a watchful eye on.

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 principles are essential: tasks should be (1) user-centric, reflecting realistic information needs, (2) dynamic, requiring up-to-date information beyond parametric knowledge, (3) unambiguous, ensuring consistent interpretation across users, and (4) multi-faceted and search-intensive, requiring search over numerous web sources and in-depth analysis. Existing benchmarks fall short of these principles, often focusing on narrow domains or posing ambiguous questions that hinder fair comparison. Guided by these principles, we introduce LiveResearchBench, a benchmark of 100 expert-curated tasks spanning daily life, enterprise, and academia, each requiring extensive, dynamic, real-time web search and synthesis. Built with over 1,500 hours of human labor, LiveResearchBench provides a rigorous basis for systematic evaluation. To evaluate citation-grounded long-form reports, we introduce DeepEval, a comprehensive suite covering both content- and report-level quality, including coverage, presentation, citation accuracy and association, consistency and depth of analysis. DeepEval integrates four complementary evaluation protocols, each designed to ensure stable assessment and high agreement with human judgments. Using LiveResearchBench and DeepEval, we conduct a comprehensive evaluation of 17 frontier deep research systems, including single-agent web search, single-agent deep research, and multi-agent systems. Our analysis reveals current strengths, recurring failure modes, and key system components needed to advance reliable, insightful deep research. Our code is available at: https://github.com/SalesforceAIResearch/LiveResearchBench.

Large Language Models increasingly possess capabilities that carry dual-use risks. While data filtering has emerged as a pretraining-time mitigation, it faces significant challenges: labeling whether data is harmful is expensive at scale, and given improving sample efficiency with larger models, even small amounts of mislabeled content could give rise to dangerous capabilities. To address risks associated with mislabeled harmful content, prior work proposed Gradient Routing (Cloud et al., 2024) -- a technique that localizes target knowledge into a dedicated subset of model parameters so they can later be removed. We explore an improved variant of Gradient Routing, which we call Selective GradienT Masking (SGTM), with particular focus on evaluating its robustness to label noise. SGTM zero-masks selected gradients such that target domain examples only update their dedicated parameters. We test SGTM's effectiveness in two applications: removing knowledge of one language from a model trained on a bilingual synthetic dataset, and removing biology knowledge from a model trained on English Wikipedia. In both cases SGTM provides better retain/forget trade-off in the presence of labeling errors compared to both data filtering and a previously proposed instantiation of Gradient Routing. Unlike shallow unlearning approaches that can be quickly undone through fine-tuning, SGTM exhibits strong robustness to adversarial fine-tuning, requiring seven times more fine-tuning steps to reach baseline performance on the forget set compared to a finetuning-based unlearning method (RMU). Our results suggest SGTM provides a promising pretraining-time complement to existing safety mitigations, particularly in settings where label noise is unavoidable.