Hort-YOLO: A multi-crop deep learning model with an integrated semi-automated annotation framework
Peer-Reviewed Publication
This month, we’re focusing on artificial intelligence (AI), a topic that continues to capture attention everywhere. Here, you’ll find the latest research news, insights, and discoveries shaping how AI is being developed and used across the world.
Updates every hour. Last Updated: 30-Dec-2025 06:11 ET (30-Dec-2025 11:11 GMT/UTC)
Electrocatalytic nitrate-to-ammonia conversion offers dual environmental and sustainable synthesis benefits, but achieving high efficiency with low-cost catalysts remains a major challenge. This review focuses on cobalt-based electrocatalysts, emphasizing their structural engineering for enhanced the performance of electrocatalytic nitrate reduction reaction (NO3RR) through dimensional control, compositional tuning, and coordination microenvironment modulation. Notably, by critically analyzing metallic cobalt, cobalt alloys, cobalt compounds, cobalt single atom and molecular catalyst configurations, we firstly establish correlations between atomic-scale structural features and catalytic performance in a coordination environment perspective for NO3RR, including the dynamic reconstruction during operation and its impact on active site. Synergizing experimental breakthroughs with computational modeling, we decode mechanisms underlying competitive hydrogen evolution suppression, intermediate adsorption-energy optimization, and durability enhancement in complex aqueous environments. The development of cobalt-based catalysts was summarized and prospected, and the emerging opportunities of machine learning in accelerating the research and development of high-performance catalysts and the configuration of series reactors for scalable nitrate-to-ammonia systems were also introduced. Bridging surface science and applications, it outlines a framework for designing multifunctional electrocatalysts to restore nitrogen cycle balance sustainably.
To enhance the electrochemical performance of lithium-ion battery anodes with higher silicon content, it is essential to engineer their microstructure for better lithium-ion transport and mitigated volume change as well. Herein, we suggest an effective approach to control the micropore structure of silicon oxide (SiOx)/artificial graphite (AG) composite electrodes using a perforated current collector. The electrode features a unique pore structure, where alternating high-porosity domains and low-porosity domains markedly reduce overall electrode resistance, leading to a 20% improvement in rate capability at a 5C-rate discharge condition. Using microstructure-resolved modeling and simulations, we demonstrate that the patterned micropore structure enhances lithium-ion transport, mitigating the electrolyte concentration gradient of lithium-ion. Additionally, perforating current collector with a chemical etching process increases the number of hydrogen bonding sites and enlarges the interface with the SiOx/AG composite electrode, significantly improving adhesion strength. This, in turn, suppresses mechanical degradation and leads to a 50% higher capacity retention. Thus, regularly arranged micropore structure enabled by the perforated current collector successfully improves both rate capability and cycle life in SiOx/AG composite electrodes, providing valuable insights into electrode engineering.
Developing effective, versatile, and high-precision sensing interfaces remains a crucial challenge in human–machine–environment interaction applications. Despite progress in interaction-oriented sensing skins, limitations remain in unit-level reconfiguration, multiaxial force and motion sensing, and robust operation across dynamically changing or irregular surfaces. Herein, we develop a reconfigurable omnidirectional triboelectric whisker sensor array (RO-TWSA) comprising multiple sensing units that integrate a triboelectric whisker structure (TWS) with an untethered hydro-sealing vacuum sucker (UHSVS), enabling reversibly portable deployment and omnidirectional perception across diverse surfaces. Using a simple dual-triangular electrode layout paired with MXene/silicone nanocomposite dielectric layer, the sensor unit achieves precise omnidirectional force and motion sensing with a detection threshold as low as 0.024 N and an angular resolution of 5°, while the UHSVS provides reliable and reversible multi-surface anchoring for the sensor units by involving a newly designed hydrogel combining high mechanical robustness and superior water absorption. Extensive experiments demonstrate the effectiveness of RO-TWSA across various interactive scenarios, including teleoperation, tactile diagnostics, and robotic autonomous exploration. Overall, RO-TWSA presents a versatile and high-resolution tactile interface, offering new avenues for intelligent perception and interaction in complex real-world environments.
Macaques can tap along to a musical beat, according to a new study – findings that upend the assumption that only animals with vocal-learning abilities can find and move to a groove. According to the authors, the discovery offers fresh insights that suggest the roots of rhythm may run far deeper in our evolutionary past than previously believed. Humans have a unique ability to perceive and move in time to a steady musical beat. It is a skill that develops early in life and requires complex pattern recognition, prediction, and motor coordination. Outside of humans, the ability to synchronize movement to rhythm – isochronicity – is strikingly rare in the animal kingdom and has only been observed in some birds and exceptional individuals of other species, leaving a gap in our understanding of its evolutionary and neurobiological roots. One powerful leading theory, the vocal-learning hypothesis, suggests that rhythmic synchronization depends on specialized brain circuits that tightly link hearing and movement, which evolved to support complex vocal learning. However, previous research shows that macaques, despite not being vocal learners, can be trained to synchronize their taps predictively with metronome beats, hinting at the neural dynamics required for isochronicity.
In this study, Vani Rajendran and colleagues investigated whether macaques trained to synchronize their taps with metronome beats could extend their metronome-tapping skills to real music in all its acoustic complexity. Rajendran et al. observed that two metronome-trained macaques independently initiated experimental trials in which they heard one of three human-selected songs and were rewarded when they tapped in time to each song’s tempo. Remarkably, both animals developed consistent tapping rhythms across all songs, and when the authors shifted the music’s tempo, the macaques’ tapping phases shifted as well, demonstrating that they were synchronizing to musical structure rather than responding reflexively to experimental cues. This behavior was observed even when the monkeys were presented with a song they had not yet heard before and when they were no longer rewarded for tapping to the beat. According to the authors, the findings suggest that, although monkeys do not experience music as fully as humans do and require substantial training, beat perception may span a broader evolutionary continuum than previously believed; it is not just restricted to vocal-learning species. “Rajendran et al. are careful to note that the abilities they observed are not natural behaviors: They were conditioned through extrinsic rewards, not the seemingly intrinsic ones that humans experience when they follow rhythmic beats,” write Asif Ghazanfar and Gavin Steingo in a related Perspective that highlights the study’s caveats. “A behavior that has been conditioned may not be equivalent to a behavior that emerges spontaneously.”
Podcast: A segment of Science's weekly podcast with Vani Rajendran, related to this research, will be available on the Science.org podcast landing page [http://www.science.org/podcasts] after the embargo lifts. Reporters are free to make use of the segments for broadcast purposes and/or quote from them – with appropriate attribution (i.e., cite "Science podcast"). Please note that the file itself should not be posted to any other Web site.
A web-based method developed by a Stanford-led team was shown to mitigate political polarization on X, the platform formerly known as Twitter, by nudging antidemocratic and extremely negative partisan posts lower in a user’s feed. The tool, which is independent of the platform, has the potential to give users more say over what they see on social media.