Newest Papers
Can online behaviors be linked to mental health? Active versus passive social network usage on depression via envy and self-esteem
Unknown Author
sciencedirect.com·2025
No abstract available
★ 4.0 (1)
View paper →Harnessing plant agriculture to mitigate climate change: A framework to evaluate synthetic biology (and other) interventions
Claudia E Vickers, Philipp Zerbe
Plant Physiology·2025
<jats:title>Abstract</jats:title>
<jats:p>Plant agriculture contributes substantially to global greenhouse gas emissions, yet it also offers powerful opportunities for climate change mitigation. Here, we focus on how to identify and prioritize synthetic biology strategies to reduce emissions and sequester carbon through plant-based interventions. Effective solutions must process large volumes of carbon, be scalable, yield a positive life-cycle balance, and be economically viable, technically feasible, and deployable in field conditions without undue damage to what remains of nature on Earth. Using Fermi estimation, we quantify the per-hectare, annual, and 100-year CO2-equivalent (CO2e) drawdown potential of emerging synthetic biology strategies—including improved CO2 fixation, reduced yield losses, root-deposited biopolymers, engineered nitrogen fixation, and methane reduction—and benchmark them against nonengineered approaches such as biochar, forestation, and fast-growing biomass crops. We used a 100-year horizon to allow for both development and implementation of high-risk but high-impact synthetic biology strategies. We integrate factors such as per-hectare effectiveness, year-on-year sequestration, deployment area, and storage durability. We demonstrate that while per-hectare impacts vary by orders of magnitude (&lt;1 to &gt;30 t CO2e/ha/year), deployment scale is the dominant factor determining total impact. Targeted synthetic biology strategies implemented across existing agricultural systems could deliver ∼120 Gt CO2e drawdown over a century and contribute to an additional ∼140 Gt CO2e drawdown. Decreasing synthetic nitrogen fertilizer use and biochar implementation have the biggest CO2e impact potential. Early-stage quantitative evaluation is critical to guide R&D toward climate-relevant solutions and deliver a prioritized portfolio of near- and long-term strategies. A transdisciplinary approach—linking synthetic biology, agronomy, engineering, and social systems—is essential to realize impact. This work offers a framework for evaluating plant agriculture-based climate mitigation strategies and highlights a key role for synthetic biology in mitigation pathways. Regular re-evaluation of strategies should be performed to ensure that they are meaningful for climate change mitigation as other factors evolve.</jats:p>
No ratings yet
View paper →Coasean Bargaining at Scale - Cosmos Institute
Cosmos Institute, Seb Krier
Substack·2025
Decentralization, coordination, and co-existence with AGI
No ratings yet
View paper →Antikythera Diffusion
Antikythera
Substack·2025
“Utopia or oblivion. It will be touch and go until the last moment” — Buckminster Fuller
No ratings yet
View paper →Faith, Fantasy, & the Female Knight
iammadeofmemories
Substack·2025
Exploring how religion, mythology, and history shape female warrior figures in fantasy, with Joan of Arc as a central archetype.
No ratings yet
View paper →Universal Complexity Bounds for Universal Gradient Methods in Nonlinear Optimization
Yurii Nesterov
arXiv·2025
In this paper, we provide the universal first-order methods of Composite Optimization with new complexity analysis. It delivers some universal convergence guarantees, which are not linked directly to any parametric problem class. However, they can be easily transformed into the rates of convergence for the particular problem classes by substituting the corresponding upper estimates for the Global Curvature Bound of the objective function. We analyze in this way the simple gradient method for nonconvex minimization, gradient methods for convex composite optimization, and their accelerated variant. For them, the only input parameter is the required accuracy of the approximate solution. The accelerated variant of our scheme automatically ensures the best possible rate of convergence simultaneously for all parametric problem classes containing the smooth part of the objective function.
★ 4.0 (1)
View paper →Technology and labor markets: Past, present, and future | Brookings
Bryan Seegmiller, Dimitris Papanikolaou, Huben Liu, Lawrence D. W. Schmidt
Brookings·2025
No abstract available
No ratings yet
View paper →Failing to Understand the Exponential, Again
Julian Schrittwieser
julian.ac·2025
Posts and writings by Julian Schrittwieser
No ratings yet
View paper →Economics and Transformative AI | Tom Cunningham – Tom Cunningham
Tom Cunningham
tecunningham.github.io·2025
Tom Cunningham blog
No ratings yet
View paper →Why are embeddings so cheap?
Piotr Mazurek
Substack·2025
or a lesson in profiling and FLOPS per dollar
No ratings yet
View paper →Thinking Augmented Pre-training
Liang Wang, Nan Yang, Shaohan Huang, Li Dong, Furu Wei
arXiv·2025
This paper introduces a simple and scalable approach to improve the data efficiency of large language model (LLM) training by augmenting existing text data with thinking trajectories. The compute for pre-training LLMs has been growing at an unprecedented rate, while the availability of high-quality data remains limited. Consequently, maximizing the utility of available data constitutes a significant research challenge. A primary impediment is that certain high-quality tokens are difficult to learn given a fixed model capacity, as the underlying rationale for a single token can be exceptionally complex and deep. To address this issue, we propose Thinking augmented Pre-Training (TPT), a universal methodology that augments text with automatically generated thinking trajectories. Such augmentation effectively increases the volume of the training data and makes high-quality tokens more learnable through step-by-step reasoning and decomposition. We apply TPT across diverse training configurations up to $100$B tokens, encompassing pre-training with both constrained and abundant data, as well as mid-training from strong open-source checkpoints. Experimental results indicate that our method substantially improves the performance of LLMs across various model sizes and families. Notably, TPT enhances the data efficiency of LLM pre-training by a factor of $3$. For a $3$B parameter model, it improves the post-training performance by over $10\%$ on several challenging reasoning benchmarks.
No ratings yet
View paper →What AI Might Mean For Workers: A Discussion - American Compass
Oren Cass
American Compass·2025
No abstract available
No ratings yet
View paper →