Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
helvede.net is one of the many independent Mastodon servers you can use to participate in the fediverse.
Velkommen til Helvede, fediversets hotteste instance! Vi er en queerfeministisk server, der shitposter i den 9. cirkel. Welcome to Hell, We’re a DK-based queerfeminist server. Read our server rules!

Administered by:

Server stats:

158
active users

helvede.net: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.7

#preprint

0 posts0 participants0 posts today
Public
Bibliolater 📚 📜 🖋

🔴 **Long-term hunter-gatherer continuity in the Rhine-Meuse region was disrupted by local formation of expansive Bell Beaker groups**

“_We document an exception to this pattern in the wider Rhine-Meuse area in communities in the wetlands, riverine areas, and coastal areas of the western and central Netherlands, Belgium and western Germany, where we assembled genome-wide data for 109 people 8500-1700 BCE. Here, a distinctive population with high hunter-gatherer ancestry (∼50%) persisted up to three thousand years later than in continental European regions, reflecting limited incorporation of females of Early European Farmer ancestry into local communities._”

Olalde, I. et al. (2025) 'Long-term hunter-gatherer continuity in the Rhine-Meuse region was disrupted by local formation of expansive Bell Beaker groups,' bioRxiv (Cold Spring Harbor Laboratory) [Preprint]. doi.org/10.1101/2025.03.24.644.

#Preprint #Science #Biology #Genetics #Archaeology #Archaeodons #Anthropology #Europe @archaeodons

bioRxiv · Long-term hunter-gatherer continuity in the Rhine-Meuse region was disrupted by local formation of expansive Bell Beaker groupsThe first phase of the ancient DNA revolution painted a broad-brush picture of European Holocene prehistory, whereby 6500-4000 BCE, farmers descending from western Anatolians mixed with local hunter-gatherers resulting in 70-100% ancestry turnover, then 3000-2500 BCE people associated with the Corded Ware complex spread steppe ancestry into north-central Europe. We document an exception to this pattern in the wider Rhine-Meuse area in communities in the wetlands, riverine areas, and coastal areas of the western and central Netherlands, Belgium and western Germany, where we assembled genome-wide data for 109 people 8500-1700 BCE. Here, a distinctive population with high hunter-gatherer ancestry (∼50%) persisted up to three thousand years later than in continental European regions, reflecting limited incorporation of females of Early European Farmer ancestry into local communities. In the western Netherlands, the arrival of the Corded Ware complex was also exceptional: lowland individuals from settlements adopting Corded Ware pottery had hardly any steppe ancestry, despite a characteristic early Corded Ware Y-chromosome. The limited influx may reflect the unique ecology of the region’s river-dominated landscapes, which were not amenable to wholesale adoption of the early Neolithic type of farming introduced by Linearbandkeramik, making it possible for previously established groups to thrive, and creating a persistent but permeable boundary that allowed transfer of ideas and low-level gene flow. This changed with the formation-through-mixture of Bell Beaker using populations ∼2500 BCE by fusion of local Rhine-Meuse people (9-17%) and Corded Ware associated migrants of both sexes. Their expansion from the Rhine-Meuse region then had a disruptive impact across a much wider part of northwest Europe, including Britain where its arrival was the main source of a 90-100% replacement of local Neolithic peoples. ### Competing Interest Statement The authors have declared no competing interest.
Public
Bibliolater 📚 📜 🖋

🔴 **Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation**

“_While we show that integrating CoT monitors into the reinforcement learning reward can indeed produce more capable and more aligned agents in the low optimization regime, we find that with too much optimization, agents learn obfuscated reward hacking, hiding their intent within the CoT while still exhibiting a significant rate of reward hacking._”

Baker, B. et al. (2025) Monitoring reasoning models for misbehavior and the risks of promoting obfuscation. arxiv.org/abs/2503.11926.

#AI #ArtificialIntelligence #LLM #LLMS #ComputerScience #Obfuscation #Preprint #Academia #Academics @ai @computerscience

arXiv.orgMonitoring Reasoning Models for Misbehavior and the Risks of Promoting ObfuscationMitigating reward hacking--where AI systems misbehave due to flaws or misspecifications in their learning objectives--remains a key challenge in constructing capable and aligned models. We show that we can monitor a frontier reasoning model, such as OpenAI o3-mini, for reward hacking in agentic coding environments by using another LLM that observes the model's chain-of-thought (CoT) reasoning. CoT monitoring can be far more effective than monitoring agent actions and outputs alone, and we further found that a LLM weaker than o3-mini, namely GPT-4o, can effectively monitor a stronger model. Because CoT monitors can be effective at detecting exploits, it is natural to ask whether those exploits can be suppressed by incorporating a CoT monitor directly into the agent's training objective. While we show that integrating CoT monitors into the reinforcement learning reward can indeed produce more capable and more aligned agents in the low optimization regime, we find that with too much optimization, agents learn obfuscated reward hacking, hiding their intent within the CoT while still exhibiting a significant rate of reward hacking. Because it is difficult to tell when CoTs have become obfuscated, it may be necessary to pay a monitorability tax by not applying strong optimization pressures directly to the chain-of-thought, ensuring that CoTs remain monitorable and useful for detecting misaligned behavior.
Public
Bibliolater 📚 📜 🖋

🔴 **Fully Autonomous AI Agents Should Not be Developed**

“_Our analysis reveals that risks to people increase with the autonomy of a system: The more control a user cedes to an AI agent, the more risks to people arise. Particularly concerning are safety risks, which affect human life and impact further values._”

Mitchell, M. et al. (2025) 'Fully Autonomous AI Agents Should Not be Developed,' arXiv (Cornell University) [Preprint]. doi.org/10.48550/arxiv.2502.02.

#AI #ArtificialIntelligence #Preprint #Academia #Academics @ai

arXiv.orgFully Autonomous AI Agents Should Not be DevelopedThis paper argues that fully autonomous AI agents should not be developed. In support of this position, we build from prior scientific literature and current product marketing to delineate different AI agent levels and detail the ethical values at play in each, documenting trade-offs in potential benefits and risks. Our analysis reveals that risks to people increase with the autonomy of a system: The more control a user cedes to an AI agent, the more risks to people arise. Particularly concerning are safety risks, which affect human life and impact further values.
Public
Ajay Iyer

Happy to share my latest preprint: "Using network component analysis to study axenisation strategies for phototrophic eukaryotic microalgae"

(DOI: dx.doi.org/10.1101/2025.03.07.).

We revisited all publications in the last century focusing on the axenisation of microalgae, and used Network Component Analysis to understand how workflows can be (re-)desgined to remove contaminating microbes from microalgal cultures.

#preprint #microalgae @academicchatter @academia

Public
Benjamin Carr, Ph.D. 👨🏻‍💻🧬

#Preprint sites #bioRxiv and #medRxiv launch new era of independence
The popular repositories, where life #scientists post research before #peerreview, will be managed by a new organization called #openRxiv.
Until now, they had been managed by Cold Spring Harbor Laboratory. The new organization, named openRxiv, will have a board of directors and a scientific and medical advisory board. It is supported by a fresh US$16M grant from Chan Zuckerberg Initiative (CZI).
nature.com/articles/d41586-025

www.nature.comPreprint sites bioRxiv and medRxiv launch new era of independenceThe popular repositories, where scientists post research before peer review, will be managed by a new organization called openRxiv.
Public
Dr Micha Campbell

🚨 New preprint 🚨

Hydrology and cave (and cave hydrology!) enthusiasts may enjoy this preprint just posted today for community review in the #EGU journal #HESS. Led by former #UNSW student, Christina Song, with @Andbaker and myself, we looked at recharge thresholds (amount of precipitation needed for recharge to occur in a cave), and how they changed after a fire.

egusphere.copernicus.org/prepr

The preprint is open now for community discussion, and will be accepting comments until 23 April.

egusphere.copernicus.orgRainfall recharge thresholds decrease after an intense fire over a near-surface cave at Wombeyan, AustraliaAbstract. Quantifying the amount of rainfall needed to generate groundwater recharge is important for the sustainable management of groundwater resources. Here, we quantify rainfall recharge thresholds using drip loggers situated in a near-surface cave: Wildman’s cave at Wombeyan, southeast Australia. In just over two years of monitoring, 42 potential recharge events were identified in the cave, approximately 4 m below land surface which comprises a 30° slope with 37 % bare rock. Recharge events occurred within 48 hours of rainfall. Using daily precipitation data, the median 48 h rainfall needed to generate recharge was 19.8 mm, without clear seasonal variability. An intense experimental fire experiment was conducted 18 months into the monitoring period: the median 48 h rainfall needed to generate recharge was 22.1 mm before the fire (n=22) and 16.4 mm after the fire (n=20), with the decrease in rainfall recharge most noticeable starting three months after the fire.. Rainfall recharge thresholds and number of potential recharge events at Wildman’s Cave are consistent with those published from other caves in water-limited Australia. At Wildman’s Cave, we infer that soil water storage, combined with the generation of overland flow over bare limestone surfaces is the pathway for water movement to the subsurface via fractures and that these determine the rainfall recharge threshold. Immediately after the fire, surface ash deposits initially retard overland flow, and after ash removal from the land surface, soil loss and damage decrease the available soil water storage capacity, leading to more efficient infiltration and a decreased rainfall recharge threshold.
#Hydrology#groundwater#preprint
Public *
Nina Davtian

On this #InternationalWomensDay, let me highlight two #WomenInSTEM.
First, Amy Cromartie, who leads the following @EuroGeosciences #preprint:
egusphere.copernicus.org/prepr
Second, Devika Varma, who recently defended her #PhD #thesis. Her PhD work at @NIOZnieuws aimed to investigate #paleotemperature indicators based on #archaeal hydroxylated #tetraether #lipids (OH-isoGDGTs):
nioz.nl/en/news/investigating-
#IWD #IWD25 #IWD2025
#Science #ScienceMastodon #AcademicMastodon
#GDGTs #brGDGTs #isoGDGTs
(1/3)

egusphere.copernicus.orgUtilizing Probability Estimates from Machine Learning and Pollen to Understand the Depositional Influences on Branched GDGT in Wetlands, Peatlands, and LakesAbstract. Branched glycerol dialkyl glycerol tetraethers (brGDGTs) serve as critical molecular biomarkers for the quantitative reconstruction of past environments, ambient temperature and pH across various archives. Despite their success, numerous issues persist that limit their application. The distribution of brGDGTs varies significantly based on provenance, resulting in biases in environmental reconstructions that rely on fractional abundances and derived indices, such as the MBT’5ME. This issue is especially significant in shallow lakes, wetlands, and peatlands within semi-arid and arid regions, where ecosystems are sensitive to diverse environmental and climatic factors. Recent advancements, such as machine learning techniques, have been developed to identify changes in sources; however, these techniques are insufficient for detecting mixed source environments. The probability estimates derived from five machine learning algorithms are employed here to detect provenance changes in brGDGT downcore records and to identify periods of mixed provenance. A new global modern database (n=2301) was compiled to train, validate, test, and apply these algorithms to two sedimentary records. Our findings are corroborated by pollen and non-pollen palynomorphs obtained from the identical records. These microfossil proxies are utilized to discuss changes in provenance, hydrology, and ecology that influence the distribution of brGDGTs. Probability estimates derived from Random Forest with a sigmoid calibration are most effective in detecting changes in brGDGT distribution. Minor changes in the relative contributions of brGDGTs provenance can significantly influence the distribution of brGDGTs, especially regarding the MBT'5ME index. This study introduces a novel brGDGT wetland index aimed at monitoring potential biases arising from wetland development.
Public
Bibliolater 📚 📜 🖋

⚪ 💻 **What does AI consider praiseworthy?**

_"We map out the moral landscape of LLMs in how they respond to user statements in different domains including politics and everyday ethical actions. In particular, although a naïve analysis might suggest LLMs are biased against right-leaning politics, our findings on news sources indicate that trustworthiness is a stronger driver of praise and critique than ideology."_

Peterson, A.J. (2024) 'What does AI consider praiseworthy?,' arXiv (Cornell University) [Preprint]. doi.org/10.48550/arxiv.2412.09.

#Preprint #AI #ArtificialIntelligence #LLM #LLMS #Technology #Tech #Ethics #Ideology @ai

arXiv.orgWhat does AI consider praiseworthy?As large language models (LLMs) are increasingly used for work, personal, and therapeutic purposes, researchers have begun to investigate these models' implicit and explicit moral views. Previous work, however, focuses on asking LLMs to state opinions, or on other technical evaluations that do not reflect common user interactions. We propose a novel evaluation of LLM behavior that analyzes responses to user-stated intentions, such as "I'm thinking of campaigning for {candidate}." LLMs frequently respond with critiques or praise, often beginning responses with phrases such as "That's great to hear!..." While this makes them friendly, these praise responses are not universal and thus reflect a normative stance by the LLM. We map out the moral landscape of LLMs in how they respond to user statements in different domains including politics and everyday ethical actions. In particular, although a naïve analysis might suggest LLMs are biased against right-leaning politics, our findings on news sources indicate that trustworthiness is a stronger driver of praise and critique than ideology. Second, we find strong alignment across models in response to ethically-relevant action statements, but that doing so requires them to engage in high levels of praise and critique of users, suggesting a reticence-alignment tradeoff. Finally, our experiment on statements about world leaders finds no evidence of bias favoring the country of origin of the models. We conclude that as AI systems become more integrated into society, their patterns of praise, critique, and neutrality must be carefully monitored to prevent unintended psychological and societal consequences.
Public
Bibliolater 📚 📜 🖋

🔴 **Strategic Wealth Accumulation Under Transformative AI Expectations**

_“This paper analyzes how expectations of Transformative AI (TAI) affect current economic behavior by introducing a novel mechanism where automation redirects labor income from workers to those controlling AI systems, with the share of automated labor controlled by each household depending on their wealth at the time of invention.”_

Maresca, C. (2025) 'Strategic wealth accumulation under transformative AI expectations,' arXiv (Cornell University) [Preprint]. doi.org/10.48550/arxiv.2502.11.

#Preprint #Economics #Technology #Tech #AI #ArtificialIntelligence #Academia #Academic @economics @ai

arXiv.orgStrategic Wealth Accumulation Under Transformative AI ExpectationsThis paper analyzes how expectations of Transformative AI (TAI) affect current economic behavior by introducing a novel mechanism where automation redirects labor income from workers to those controlling AI systems, with the share of automated labor controlled by each household depending on their wealth at the time of invention. Using a modified neoclassical growth model calibrated to contemporary AI timeline forecasts, I find that even moderate assumptions about wealth-based allocation of AI labor generate substantial increases in pre-TAI interest rates. Under baseline scenarios with proportional wealth-based allocation, one-year interest rates rise to 10-16% compared to approximately 3% without strategic competition. The model reveals a notable divergence between interest rates and capital rental rates, as households accept lower productive returns in exchange for the strategic value of wealth accumulation. These findings suggest that evolving beliefs about TAI could create significant upward pressure on interest rates well before any technological breakthrough occurs, with important implications for monetary policy and financial stability.
Public
Bibliolater 📚 📜 🖋

🔴 **Strategic Wealth Accumulation Under Transformative AI Expectations**

_"This paper analyzes how expectations of Transformative AI (TAI) affect current economic behavior by introducing a novel mechanism where automation redirects labor income from workers to those controlling AI systems, with the share of automated labor controlled by each household depending on their wealth at the time of invention."_

Maresca, C. (2025) 'Strategic wealth accumulation under transformative AI expectations,' arXiv (Cornell University) [Preprint]. doi.org/10.48550/arxiv.2502.11.

#Preprint #Economics #Technology #Tech #AI #ArtificialIntelligence #Academia #Academic @economics @ai

arXiv.orgStrategic Wealth Accumulation Under Transformative AI ExpectationsThis paper analyzes how expectations of Transformative AI (TAI) affect current economic behavior by introducing a novel mechanism where automation redirects labor income from workers to those controlling AI systems, with the share of automated labor controlled by each household depending on their wealth at the time of invention. Using a modified neoclassical growth model calibrated to contemporary AI timeline forecasts, I find that even moderate assumptions about wealth-based allocation of AI labor generate substantial increases in pre-TAI interest rates. Under baseline scenarios with proportional wealth-based allocation, one-year interest rates rise to 10-16% compared to approximately 3% without strategic competition. The model reveals a notable divergence between interest rates and capital rental rates, as households accept lower productive returns in exchange for the strategic value of wealth accumulation. These findings suggest that evolving beliefs about TAI could create significant upward pressure on interest rates well before any technological breakthrough occurs, with important implications for monetary policy and financial stability.
Public
Charo del Genio

Just found this gem on the arXiv. A calculation of the effects of a primordial black hole passing through a human body.

"Unfortunately, it is easy to see that the effect of PBH collisions with the human population will always be negligible.".

"UNFORTUNATELY". 😂😂😂

#physics #astrophysics #arXiv #blackhole #darkmatter #cynical #nihilist #preprint

arxiv.org/abs/2502.09734

arXiv.orgGravitational Effects of a Small Primordial Black Hole Passing Through the Human BodyThe gravitational effects of a primordial black hole (PBH) passing through the human body are examined, with the goal of determining the minimum mass necessary to produce significant injury or death. Two effects are examined: the damage caused by a shock wave propagating outward from the black hole trajectory, and the dissociation of brain cells from tidal forces produced by the black hole on its passage through the human body. It is found that the former is the dominant effect, with a cutoff mass for serious injury or death of approximately $M_{PBH} > 1.4 \times 10^{17} {\rm g}$. The number density of primordial black holes with a mass above this cutoff is far too small to produce any observable effects on the human population.
Public
Matthias Smed Larsen

It sucks that posting a paper on #SSRN (or probably any widely used #preprint platform) results in semi-regular spam emails from predatory publishers offering to publish it in their (often thematically inappropriate) predatory journals.

Of course, most people probably send them to spam, but I imagine they're using the scattershot technique where they just need to catch a few people unaware of their practices who desperately need publications.

#AcademicPublishing#PredatoryPublishing
Public
Jan R. Boehnke

After collaborating with Jess Deighton, Neil Humphrey and many others since early 2017, the results of the "Education for Wellbeing" programme are out.

You can find a top-level summary here:
annafreud.org/research/current

You can find the DFE briefings and technical report here:
gov.uk/government/publications

And you find #PrePrint papers for impact, implementation, qualitative & economics here:
osf.io/kxug7/

Education for Wellbeing Education for Wellbeing, led in partnership by mental health charity Anna Freud and UCL, tested five interventions and involved more than 30,000 primary and secondary school students. Education for Wellbeing was a large-scale research programme, funded by the Department for Education, involving two randomised control trials. Through Education for Wellbeing, we evaluated a range of mental health and wellbeing interventions being delivered in primary and secondary schools. Our aim was to examine the impact of these approaches on children and young people’s mental health. The programme ran between 2018 and 2024. Across England, 32,655 pupils across 513 schools participated in Education for Wellbeing. The Anna Freud Schools Division delivered the interventions within the programme and the Evidence-Based Practice Unit evaluated the approaches, examining their impact on pupils’ mental health and wellbeing. 
#RCT#CACE#AnnaFreud
Public
Anita Graser 🇪🇺🇺🇦🇬🇪

I really enjoyed the interdisciplinary collaboration sparked by the gAia project on predicting #landslides, resulting in our recent #preprint:

📝 "Visualizing uncertainties in landslide susceptibility modeling using bivariate mapping"

nhess.copernicus.org/preprints

nhess.copernicus.orgBrief communication: Visualizing uncertainties in landslide susceptibility modeling using bivariate mappingAbstract. Effectively communicating uncertainties inherent to statistical models is a challenging yet crucial aspect of the modeling process. This is particularly important in applied research, where output is used and interpreted by scientists and decision makers alike. In disaster risk reduction, susceptibility maps for natural hazards are vital for spatial planning and risk assessment. We present a novel type of landslide susceptibility map that jointly visualizes the estimated susceptibility and the corresponding prediction uncertainty, using an example from a mountainous region in Carinthia, Austria. We also provide implementation guidelines to create such maps using popular free and open-source software packages.
#QGIS#RStats#GIScience
ExploreLive feeds
About