@halfwitty.bsky.social
Large Language Model Reasoning Failures
"... unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with confidence while doing the wrong thing."
www.arxiv.org/pdf/2602.061...
"... unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with confidence while doing the wrong thing."
www.arxiv.org/pdf/2602.061...
February 11, 2026 at 4:32 PM
Large Language Model Reasoning Failures
"... unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with confidence while doing the wrong thing."
www.arxiv.org/pdf/2602.061...
"... unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with confidence while doing the wrong thing."
www.arxiv.org/pdf/2602.061...
A paper in Nature presents OpenScholar, an open-source language model that can outperform commercial large language models (LLMs) in performing accurate literature reviews.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Synthesizing scientific literature with retrieval-augmented language models - Nature
A specialized, open-source, retrieval-augmented language model is introduced for answering scientific queries and synthesizing literature, the responses of which are shown to be preferred by human eva...
www.nature.com
February 6, 2026 at 11:03 PM
A paper in Nature presents OpenScholar, an open-source language model that can outperform commercial large language models (LLMs) in performing accurate literature reviews.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Researchers discover enzyme SCoR2, which promotes fat storage and lipogenesis
They found that blocking this nitric oxide-removing enzyme prevented weight gain and liver injury in mouse models. The same drug also lowered bad cholesterol.
www.science.org/doi/10.1126/...
They found that blocking this nitric oxide-removing enzyme prevented weight gain and liver injury in mouse models. The same drug also lowered bad cholesterol.
www.science.org/doi/10.1126/...
The protein denitrosylase SCoR2 regulates lipogenesis and fat storage
Removal of SNO groups from hepatocyte and adipocyte proteins induces lipid synthesis and weight gain.
www.science.org
February 3, 2026 at 8:01 PM
Researchers discover enzyme SCoR2, which promotes fat storage and lipogenesis
They found that blocking this nitric oxide-removing enzyme prevented weight gain and liver injury in mouse models. The same drug also lowered bad cholesterol.
www.science.org/doi/10.1126/...
They found that blocking this nitric oxide-removing enzyme prevented weight gain and liver injury in mouse models. The same drug also lowered bad cholesterol.
www.science.org/doi/10.1126/...
Study: Ultra-processed foods should be treated more like cigarettes than food
UPFs are made to encourage addiction and consumption and should be regulated like tobacco.
onlinelibrary.wiley.com/doi/10.1111/...
UPFs are made to encourage addiction and consumption and should be regulated like tobacco.
onlinelibrary.wiley.com/doi/10.1111/...
February 3, 2026 at 4:43 PM
Study: Ultra-processed foods should be treated more like cigarettes than food
UPFs are made to encourage addiction and consumption and should be regulated like tobacco.
onlinelibrary.wiley.com/doi/10.1111/...
UPFs are made to encourage addiction and consumption and should be regulated like tobacco.
onlinelibrary.wiley.com/doi/10.1111/...
Sub-unit-cell resolution imaging of a Wigner crystal formed by 4f electrons in a YbCl₃ monolayer reveals a highly localized, strongly correlated electron lattice with record-high density and melting temperature.
phys.org/news/2026-01...?
phys.org/news/2026-01...?
Imaging the Wigner crystal state in a new type of quantum material
In some solid materials under specific conditions, mutual Coulomb interactions shape electrons into many-body correlated states, such as Wigner crystals, which are essentially solids made of electrons...
phys.org
February 2, 2026 at 8:15 PM
Sub-unit-cell resolution imaging of a Wigner crystal formed by 4f electrons in a YbCl₃ monolayer reveals a highly localized, strongly correlated electron lattice with record-high density and melting temperature.
phys.org/news/2026-01...?
phys.org/news/2026-01...?
Research suggests that "playing the victim" does not signal weakness to voters.
Instead, politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not, making it a highly attractive strategy.
www.tandfonline.com/.../10.../10...
Instead, politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not, making it a highly attractive strategy.
www.tandfonline.com/.../10.../10...
February 2, 2026 at 8:13 PM
Research suggests that "playing the victim" does not signal weakness to voters.
Instead, politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not, making it a highly attractive strategy.
www.tandfonline.com/.../10.../10...
Instead, politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not, making it a highly attractive strategy.
www.tandfonline.com/.../10.../10...
Research suggests that "playing the victim" does not signal weakness to voters.
Politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not.
www.tandfonline.com/doi/full/10....
Politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not.
www.tandfonline.com/doi/full/10....
Claims of Victimhood Shield Politicians from Political Scandals
Political scandals can shape entire elections and political structures within society. Scholars have long pointed to different strategies politicians use to respond to such scandals, finding strate...
www.tandfonline.com
February 2, 2026 at 1:44 PM
Research suggests that "playing the victim" does not signal weakness to voters.
Politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not.
www.tandfonline.com/doi/full/10....
Politicians who emphasized their own victimhood during a scandal were often evaluated as more competent than those who did not.
www.tandfonline.com/doi/full/10....
Sub-unit-cell resolution imaging of a Wigner crystal formed by 4f electrons in a YbCl₃ monolayer reveals a highly localized, strongly correlated electron lattice with record-high density and melting temperature.
phys.org/news/2026-01...?
phys.org/news/2026-01...?
Imaging the Wigner crystal state in a new type of quantum material
In some solid materials under specific conditions, mutual Coulomb interactions shape electrons into many-body correlated states, such as Wigner crystals, which are essentially solids made of electrons...
phys.org
February 2, 2026 at 1:41 PM
Sub-unit-cell resolution imaging of a Wigner crystal formed by 4f electrons in a YbCl₃ monolayer reveals a highly localized, strongly correlated electron lattice with record-high density and melting temperature.
phys.org/news/2026-01...?
phys.org/news/2026-01...?
Researchers have developed a contactless method that analyzes a combination of patients’ heart rate data and facial expressions to estimate the pain they’re feeling.
No more “on a scale of 1 to 10”.
spectrum.ieee.org/machine-lear...
spectrum.ieee.org/machine-lear...
No more “on a scale of 1 to 10”.
spectrum.ieee.org/machine-lear...
spectrum.ieee.org/machine-lear...
AI and Cameras Team Up for Better Pain Detection
How does a camera and heart rate data team up to estimate pain levels? Find out about this innovative approach.
spectrum.ieee.org
January 28, 2026 at 10:16 PM
Researchers have developed a contactless method that analyzes a combination of patients’ heart rate data and facial expressions to estimate the pain they’re feeling.
No more “on a scale of 1 to 10”.
spectrum.ieee.org/machine-lear...
spectrum.ieee.org/machine-lear...
No more “on a scale of 1 to 10”.
spectrum.ieee.org/machine-lear...
spectrum.ieee.org/machine-lear...
U.S. government has lost more than 10,000 STEM Ph.D.s since Trump took office
A Science analysis reveals how many were fired, retired, or quit across 14 agencies
www.science.org/content/arti...
A Science analysis reveals how many were fired, retired, or quit across 14 agencies
www.science.org/content/arti...
U.S. government has lost more than 10,000 STEM Ph.D.s since Trump took office
A Science analysis reveals how many were fired, retired, or quit across 14 agencies
www.science.org
January 28, 2026 at 10:14 PM
U.S. government has lost more than 10,000 STEM Ph.D.s since Trump took office
A Science analysis reveals how many were fired, retired, or quit across 14 agencies
www.science.org/content/arti...
A Science analysis reveals how many were fired, retired, or quit across 14 agencies
www.science.org/content/arti...
Scientists engineer unsinkable metal tubes
The superhydrophobic design could lead to resilient ships, floating platforms, and renewable energy innovations.
advanced.onlinelibrary.wiley.com/doi/10.1002/...
www.rochester.edu/newscenter/u...
The superhydrophobic design could lead to resilient ships, floating platforms, and renewable energy innovations.
advanced.onlinelibrary.wiley.com/doi/10.1002/...
www.rochester.edu/newscenter/u...
Geometry‐Enabled Recoverable Floating Superhydrophobic Metallic Tubes
This study presents a geometrically optimized superhydrophobic tube that achieves unprecedented stability in aquatic environments. By keeping the inner diameter within the critical water dimple depth....
advanced.onlinelibrary.wiley.com
January 28, 2026 at 10:08 PM
Scientists engineer unsinkable metal tubes
The superhydrophobic design could lead to resilient ships, floating platforms, and renewable energy innovations.
advanced.onlinelibrary.wiley.com/doi/10.1002/...
www.rochester.edu/newscenter/u...
The superhydrophobic design could lead to resilient ships, floating platforms, and renewable energy innovations.
advanced.onlinelibrary.wiley.com/doi/10.1002/...
www.rochester.edu/newscenter/u...
Conservative college students don’t face greater barriers to campus resources
Do colleges discriminate against conservative students? New data says no.
link.springer.com/article/10.1...
Do colleges discriminate against conservative students? New data says no.
link.springer.com/article/10.1...
Political Bias in College Student Access To Campus Resources - Political Behavior
In line with the expectations of social identity theory and work on affective polarization and ideological sorting, numerous actors allege widespread bias against political conservatives in America’s ...
link.springer.com
January 28, 2026 at 9:46 PM
Conservative college students don’t face greater barriers to campus resources
Do colleges discriminate against conservative students? New data says no.
link.springer.com/article/10.1...
Do colleges discriminate against conservative students? New data says no.
link.springer.com/article/10.1...
Single Dose Oral Cholera Vaccine Completes Phase 1 Trial
www.thelancet.com/journals/lan...
www.massgeneralbrigham.org/en/about/new...
www.thelancet.com/journals/lan...
www.massgeneralbrigham.org/en/about/new...
Safety and immunogenicity of PanChol, a single-dose live-attenuated oral cholera vaccine: results from a phase 1a, double-blind, randomised, placebo-controlled trial
A single oral dose of PanChol was safe and well tolerated at all doses and induced
100% vibriocidal seroconversion over a 100 000-fold dose range. These findings support
the progression of PanChol int...
www.thelancet.com
January 28, 2026 at 9:37 PM
Single Dose Oral Cholera Vaccine Completes Phase 1 Trial
www.thelancet.com/journals/lan...
www.massgeneralbrigham.org/en/about/new...
www.thelancet.com/journals/lan...
www.massgeneralbrigham.org/en/about/new...
Researchers Beam Power From a Moving Airplane for the first time
The concept is to use less-dense, infrared waves. Existing utility-scale solar farms would be able to receive the beamed energy just like they receive the sun’s energy during daylight hours.
spectrum.ieee.org/wireless-pow...
The concept is to use less-dense, infrared waves. Existing utility-scale solar farms would be able to receive the beamed energy just like they receive the sun’s energy during daylight hours.
spectrum.ieee.org/wireless-pow...
Researchers Beam Wireless Power From a Moving Airplane
Can we beam space-based solar power back to Earth? Startup Overview Energy demonstrates wireless power beaming from a moving aircraft.
spectrum.ieee.org
January 27, 2026 at 2:07 PM
Researchers Beam Power From a Moving Airplane for the first time
The concept is to use less-dense, infrared waves. Existing utility-scale solar farms would be able to receive the beamed energy just like they receive the sun’s energy during daylight hours.
spectrum.ieee.org/wireless-pow...
The concept is to use less-dense, infrared waves. Existing utility-scale solar farms would be able to receive the beamed energy just like they receive the sun’s energy during daylight hours.
spectrum.ieee.org/wireless-pow...
Shingles vaccine may help slow biological aging, study finds
Scientists found that participants who had received the shingles vaccine had a slower overall biological aging on average compared to the unvaccinated.
www.medicalnewstoday.com/articles/shi...
study pubmed.ncbi.nlm.nih.gov/41556115/
Scientists found that participants who had received the shingles vaccine had a slower overall biological aging on average compared to the unvaccinated.
www.medicalnewstoday.com/articles/shi...
study pubmed.ncbi.nlm.nih.gov/41556115/
Longevity: Can the shingles vaccine help slow down aging?
Older adults vaccinated against shingles appear to experience slower biological aging, according to a new study.
www.medicalnewstoday.com
January 26, 2026 at 4:28 PM
Shingles vaccine may help slow biological aging, study finds
Scientists found that participants who had received the shingles vaccine had a slower overall biological aging on average compared to the unvaccinated.
www.medicalnewstoday.com/articles/shi...
study pubmed.ncbi.nlm.nih.gov/41556115/
Scientists found that participants who had received the shingles vaccine had a slower overall biological aging on average compared to the unvaccinated.
www.medicalnewstoday.com/articles/shi...
study pubmed.ncbi.nlm.nih.gov/41556115/
High body mass index identified as a direct cause of vascular dementia.
A new genetic analysis indicates that maintaining a healthy weight and managing hypertension could prevent a substantial number of dementia cases.
academic.oup.com/jcem/advance...
A new genetic analysis indicates that maintaining a healthy weight and managing hypertension could prevent a substantial number of dementia cases.
academic.oup.com/jcem/advance...
High Body Mass Index as a Causal Risk Factor for Vascular-Related Dementia: A Mendelian Randomization Study
AbstractContext. Obesity is associated with a high risk of vascular-related dementia with metabolic risk factors as potential mediators, but questions of c
academic.oup.com
January 26, 2026 at 4:24 PM
High body mass index identified as a direct cause of vascular dementia.
A new genetic analysis indicates that maintaining a healthy weight and managing hypertension could prevent a substantial number of dementia cases.
academic.oup.com/jcem/advance...
A new genetic analysis indicates that maintaining a healthy weight and managing hypertension could prevent a substantial number of dementia cases.
academic.oup.com/jcem/advance...
Viewing "metabolic" drugs like Ozempic/Wegovy (GLP-1s) and Jardiance (SGLT2 inhibitors) as powerful anti-inflammatory treatments.
Conditions like obesity or Type 2 Diabetes, high blood sugar, and excess fat tissue triggers chronic low-grade inflammation.
www.cell.com/trends/endoc...
Conditions like obesity or Type 2 Diabetes, high blood sugar, and excess fat tissue triggers chronic low-grade inflammation.
www.cell.com/trends/endoc...
Repurposing metabolic drugs as anti-inflammatory agents
Dysregulation of bodyweight systemic metabolism is intrinsically linked to an inflammatory
phenotype, with each underpinning the other. Over the past decade, new classes of
drug, such as glucagon-like...
www.cell.com
January 23, 2026 at 9:19 PM
Viewing "metabolic" drugs like Ozempic/Wegovy (GLP-1s) and Jardiance (SGLT2 inhibitors) as powerful anti-inflammatory treatments.
Conditions like obesity or Type 2 Diabetes, high blood sugar, and excess fat tissue triggers chronic low-grade inflammation.
www.cell.com/trends/endoc...
Conditions like obesity or Type 2 Diabetes, high blood sugar, and excess fat tissue triggers chronic low-grade inflammation.
www.cell.com/trends/endoc...
The secrets of baseball's magic mud: Study quantifies its properties to show it's not simply a superstition
They found that it is composed of nonswelling clay, which
reduces shine without being slippery, and quartz, which helps to abrade the leather for grip.
www.pnas.org/doi/10.1073/...
They found that it is composed of nonswelling clay, which
reduces shine without being slippery, and quartz, which helps to abrade the leather for grip.
www.pnas.org/doi/10.1073/...
January 23, 2026 at 7:52 PM
The secrets of baseball's magic mud: Study quantifies its properties to show it's not simply a superstition
They found that it is composed of nonswelling clay, which
reduces shine without being slippery, and quartz, which helps to abrade the leather for grip.
www.pnas.org/doi/10.1073/...
They found that it is composed of nonswelling clay, which
reduces shine without being slippery, and quartz, which helps to abrade the leather for grip.
www.pnas.org/doi/10.1073/...
Researchers discover potential new target to treat Parkinson’s disease
These findings highlight the αSyn–ClpP interaction as a druggable target and support CS2 as a potential disease-modifying therapy for PD and related synucleinopathies.
case.edu/news/researc...
These findings highlight the αSyn–ClpP interaction as a druggable target and support CS2 as a potential disease-modifying therapy for PD and related synucleinopathies.
case.edu/news/researc...
Researchers discover potential new target to treat Parkinson’s disease | CWRU Newsroom | Case Western Reserve University
About 1 million Americans suffer from Parkinson’s disease, with around 90,000 new cases diagnosed each year, according to the Parkinson’s Foundati...
case.edu
January 23, 2026 at 2:50 PM
Researchers discover potential new target to treat Parkinson’s disease
These findings highlight the αSyn–ClpP interaction as a druggable target and support CS2 as a potential disease-modifying therapy for PD and related synucleinopathies.
case.edu/news/researc...
These findings highlight the αSyn–ClpP interaction as a druggable target and support CS2 as a potential disease-modifying therapy for PD and related synucleinopathies.
case.edu/news/researc...
Scientists sequence a woolly rhino genome from a 14,400-year-old wolf’s stomach
Genome Shows no Recent Inbreeding in Near-Extinction Woolly Rhinoceros Sample Found in Ancient Wolf's Stomach
arstechnica.com/science/2026...
Genome Shows no Recent Inbreeding in Near-Extinction Woolly Rhinoceros Sample Found in Ancient Wolf's Stomach
arstechnica.com/science/2026...
Scientists sequence a woolly rhino genome from a 14,400-year-old wolf’s stomach
Fortunately for paleogeneticists, wolf puppies don't chew their food thoroughly.
arstechnica.com
January 20, 2026 at 10:25 PM
Scientists sequence a woolly rhino genome from a 14,400-year-old wolf’s stomach
Genome Shows no Recent Inbreeding in Near-Extinction Woolly Rhinoceros Sample Found in Ancient Wolf's Stomach
arstechnica.com/science/2026...
Genome Shows no Recent Inbreeding in Near-Extinction Woolly Rhinoceros Sample Found in Ancient Wolf's Stomach
arstechnica.com/science/2026...
Cannabis Provides Sustained Relief for Patients With Treatment-Resistant Depression
Patients suffering from treatment-resistant depression report sustained improvements in their health-related quality of life following the use of medical cannabis preparations
www.sciencedirect.com/science/arti...
Patients suffering from treatment-resistant depression report sustained improvements in their health-related quality of life following the use of medical cannabis preparations
www.sciencedirect.com/science/arti...
UK Medical Cannabis Registry: A two-year case series of clinical outcomes in depression
Whilst preclinical evidence details the effects of cannabinoids on mood regulation, there is a paucity of clinical evidence on the use of cannabis-bas…
www.sciencedirect.com
January 20, 2026 at 10:20 PM
Cannabis Provides Sustained Relief for Patients With Treatment-Resistant Depression
Patients suffering from treatment-resistant depression report sustained improvements in their health-related quality of life following the use of medical cannabis preparations
www.sciencedirect.com/science/arti...
Patients suffering from treatment-resistant depression report sustained improvements in their health-related quality of life following the use of medical cannabis preparations
www.sciencedirect.com/science/arti...
Study Finds Alpha-2 Receptor Drugs Reduce Heavy Alcohol Drinking
Findings point to alpha-2 adrenergic receptors as a key modulator of heavy drinking and alcohol-related cognitive problems.
www.bumc.bu.edu/camed/news-e...
Findings point to alpha-2 adrenergic receptors as a key modulator of heavy drinking and alcohol-related cognitive problems.
www.bumc.bu.edu/camed/news-e...
January 20, 2026 at 10:17 PM
Study Finds Alpha-2 Receptor Drugs Reduce Heavy Alcohol Drinking
Findings point to alpha-2 adrenergic receptors as a key modulator of heavy drinking and alcohol-related cognitive problems.
www.bumc.bu.edu/camed/news-e...
Findings point to alpha-2 adrenergic receptors as a key modulator of heavy drinking and alcohol-related cognitive problems.
www.bumc.bu.edu/camed/news-e...
Study: Trump supporters and insecure men more likely to value a large penis
The penis serves as a symbol of status and dominance. The desire for a larger one is driven by feelings of humiliation to meet social expectations of manhood.
www.psypost.org/trump-suppor...
psycnet.apa.org/fulltext/202...
The penis serves as a symbol of status and dominance. The desire for a larger one is driven by feelings of humiliation to meet social expectations of manhood.
www.psypost.org/trump-suppor...
psycnet.apa.org/fulltext/202...
Trump supporters and insecure men more likely to value a large penis, according to new research
Research involving four studies found that men’s value for a large penis relates to threatened masculinity, dominance, and aggression. The findings suggest that feeling inadequate as a man increases t...
www.psypost.org
January 19, 2026 at 1:17 PM
Study: Trump supporters and insecure men more likely to value a large penis
The penis serves as a symbol of status and dominance. The desire for a larger one is driven by feelings of humiliation to meet social expectations of manhood.
www.psypost.org/trump-suppor...
psycnet.apa.org/fulltext/202...
The penis serves as a symbol of status and dominance. The desire for a larger one is driven by feelings of humiliation to meet social expectations of manhood.
www.psypost.org/trump-suppor...
psycnet.apa.org/fulltext/202...
Fermented Lacticaseibacillus Paracasei Cultures Ameliorate Colitis by Modulating Microbiota-Derived Tryptophan Metabolism and Macrophage Polarization
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
January 19, 2026 at 1:11 PM
Fermented Lacticaseibacillus Paracasei Cultures Ameliorate Colitis by Modulating Microbiota-Derived Tryptophan Metabolism and Macrophage Polarization
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
advanced.onlinelibrary.wiley.com/doi/10.1002/...
A systematic review of 43 studies confirms prenatal acetaminophen does not increase autism, ADHD, or intellectual disability risks.
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...
Prenatal paracetamol exposure and child neurodevelopment: a systematic review and meta-analysis
Current evidence does not indicate a clinically important increase in the likelihood
of autism spectrum disorder, ADHD, or intellectual disability in children of pregnant
individuals who use paracetam...
www.thelancet.com
January 19, 2026 at 1:06 PM
A systematic review of 43 studies confirms prenatal acetaminophen does not increase autism, ADHD, or intellectual disability risks.
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...
www.thelancet.com/journals/lan...