The Sudanese government's allegations that Ethiopia is behind a deadly wave of unmanned aerial assaults belie a deep and longstanding rift between the two countries, which threatens to intensify internal conflicts in both. Sudan has officially lodged complaints with the United Nations Security Council and the African Union's Peace and Security Council over allegations that the Rapid Support Forces (RSF) paramilitary - which has been at war with the Sudanese Armed Forces (SAF) since April 2023 - is using Ethiopian territory to launch attacks against civilian sites, including power stations. "The mutual accusations are many and almost daily, and the tension on the part of the Sudanese army has reached high levels, to the degree that made some voices inside the Sudanese army and the forces allied with it demand that Sudan declare war on Ethiopia based on these interventions," Al-Almin told Al Jazeera. Suspected RSF drone attacks have heavily damaged civilian infrastructure in SAF-controlled areas. More than 90 percent of power stations in Sudanese cities have been rendered out of service due to drone attacks, and al-Almin said the technological capabilities of these weapons indicate significant external backing.

Fighting and drone strikes in northern Ethiopia are pushing the country's fragile health system towards collapse, the head of the World Health Organization (WHO) has warned. Director-General Tedros Adhanom Ghebreyesus reported in a statement released on Monday that the conflict between rebels from the northern Tigray region and the government in Addis Ababa had killed several civilians in recent days and strained medical facilities. At least 15 people, several of them children, died in a series of drone strikes that hit a camp for displaced civilians in Ethiopia's Amhara region on Friday, sources involved in a rescue operation told the Reuters news agency. Tedros, the world's highest-profile Tigrayan, confirmed that children were among those reported to be killed at the Jaari 2 displacement camp in Amhara. In Rama, in neighbouring Tigray, several people were killed while seeking food.

Renewed fighting in Ethiopia's northern Tigray region has raised fears of a wider regional conflict. In Adigrat, a city close to the Eritrean border, residents recall recent drone strikes by Ethiopian forces against Eritrean troops. Why the US is holding secret talks with Myanmar's military ruler Why is Trump now encouraging mail-in voting?

U.S. Embassy in Ethiopia warns of long-range attacks on military installations as analysts say the Horn of Africa could erupt in a new war with regional backing.

Rebel forces in Ethiopia's northern Tigray region have said they have temporarily withdrawn from the regional capital, Mekelle, as fighting with the national army moves towards the city. The Tigray People's Liberation Front (TPLF), which has been running the region, said its decision to relocate was made because of concerns about risks to civilians. Two residents in Mekelle told the BBC that pro-government forces now control the city and TPLF forces have abandoned it. One resident also said mobile phone services appear to have resumed in parts of the city. Fighting has erupted in recent weeks between the TPLF and federal government forces alongside allied fighters in parts of Ethiopia's northern regions of Tigray, Afar and Amhara.

Explosions have rung out in Addis Ababa as fighting between Ethiopia's federal forces and a coalition of armed opposition groups widens in the country's north. Three diplomatic sources told the Reuters news agency on Thursday that the blasts were heard overnight in the Ethiopian capital. The blasts came hours after Addis Ababa instituted a ban on drone-flying in the capital and surrounding areas. The alliance unites seven armed groups, including the Tigray People's Liberation Front (TPLF) and groups from the Amhara, Afar, Somali and Benishangul-Gumuz regions. It says it wants to remove Prime Minister Abiy Ahmed's government.

Share Ethiopia's Gobena, Kenya's Olympic champion Jepchirchir win Sydney Marathon on social media Ethiopia's Addisu Gobena and Kenyan world champion Peres Jepchirchir surged to commanding Sydney Marathon wins in the men's and women's races . It was the first victory in a World Marathon Major for 21-year-old Gobena, who came second last year, while Jepchirchir, in Australia on Sunday, added to her previous triumphs in Boston, New York and London. The dominant run shattered the course record by almost one and a half minutes, with last year's winner Hailemaryam Kiros dropping from the pace in the dying stages. "I am very happy," said Gobena, who was a javelin thrower before transitioning to long-distance running, gaining international attention by winning the 2024 Dubai Marathon. "Last year I came second, I learnt from my mistake and I went back home and I worked on my mistake to perform and break the record today." The unstoppable Jepchirchir was in a league of her own, crossing the finish line in front of the Sydney Opera House in 2:18:31, nearly four minutes ahead of fellow Kenyan Irine Cheptai and Ethiopia's Shure Demise Ware.

Kenya's Sabastian Sawe has become the first man to run a marathon in under two hours, winning the London Marathon in 1:59:30. Ethiopia's Tigst Assefa defended her London Marathon crown on Sunday, breaking her own world record. The 31-year-old, who has never lost a marathon, smashed the world record by 65 seconds. Yomif Kejelcha of Ethiopia stayed on Sawe's heels for most of the 42.195km course before fading down the final stretch to take second in his marathon debut with 1:59:41, while Jacob Kiplimo of Uganda won bronze in 2:02:28. All three finished under Kiptum's previous record time.

What are the limits of controlling language models via synthetic training data? We develop a reinforcement learning (RL) primitive, the Dataset Policy Gradient (DPG), which can precisely optimize synthetic data generators to produce a dataset of targeted examples. When used for supervised fine-tuning (SFT) of a target model, these examples cause the target model to do well on a differentiable metric of our choice. Our approach achieves this by taking exact data attribution via higher-order gradients and using those scores as policy gradient rewards. We prove that this procedure closely approximates the true, intractable gradient for the synthetic data generator. To illustrate the potential of DPG, we show that, using only SFT on generated examples, we can cause the target model's LM head weights to (1) embed a QR code, (2) embed the pattern $\texttt{67}$, and (3) have lower $\ell^2$ norm. We additionally show that we can cause the generator to (4) rephrase inputs in a new language and (5) produce a specific UUID, even though neither of these objectives is conveyed in the generator's input prompts. These findings suggest that DPG is a powerful and flexible technique for shaping model properties using only synthetic training examples.

This easy-to-use, Wi-Fi-enabled bean-to-cup brewer is good, but not quite great. App is simple and works well. "Smart" features only work with Amazon Alexa and Google Assistant. Integrating with HomeKit via third-party apps is not worth the effort. Pricey for what's essentially an auto-drip machine that works with an app, which is no longer novel or futuristic.