Support Pullquote, upgrade to Pro!
(Or just tweet your Pullquote for free!)
With Pullquote Pro, you'll get to:
- share on Facebook
- schedule tweets
- tweet from multiple accounts
- edit quotes
- customize colors
- change fonts
- save and index quotes
- private quotes
Choose a plan: $5/month $50/year (includes free access to any new features)
Recent quotes:
Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial | Clinical Decision Support | JAMA Network Open | JAMA Network
An unexpected secondary result was that the LLM alone performed significantly better than both groups of humans, similar to a recent study with different LLM technology.31 This may be explained by the sensitivity of LLM output to prompt formulation.32 There are numerous frameworks for prompting LLMs and an emerging consensus on prompting strategies, many of which focus on providing details on the task, context, and instructions; our prompt was iteratively developed using these frameworks. Training clinicians in best prompting practices may improve physician performance with LLMs. Alternatively, organizations could invest in predefined prompting for diagnostic decision support integrated into clinical workflows and documentation, enabling synergy between the tools and clinicians. Prior studies on AI systems show disparate effects depending on the component of the diagnostic process they are used in.33,34 Given the conversational nature of chatbots, changes in how the LLM interacts with humans, for example by specifically pointing out features that do not fit the differential diagnosis, might improve diagnostic and reflective performance.35,36 More generally, we see opportunity with deliberate consideration and redesign of medical education and practice frameworks that adapt to disruptive emerging technologies and enable the best use of computer and human resources to deliver optimal medical care.
ChatGPT Defeated Doctors at Diagnosing Illness - The New York Times
The chatbot, from the company OpenAI, scored an average of 90 percent when diagnosing a medical condition from a case report and explaining its reasoning. Doctors randomly assigned to use the chatbot got an average score of 76 percent. Those randomly assigned not to use it had an average score of 74 percent.
The study showed more than just the chatbot’s superior performance.
It unveiled doctors’ sometimes unwavering belief in a diagnosis they made, even when a chatbot potentially suggests a better one.
Polymarket Whale Traders: How 1% of Bettors Are Boosting Trump’s Odds - Bloomberg
Among this subset of users, 16% of bets can be traced to just 10 whales. On Oct. 17, six of these whales drove 37% of predictions. In one 12-hour window, they placed $5.2 million in bets as Trump’s odds rose about four percentage points.
Financial Statement Analysis with Large Language Models by Alex Kim, Maximilian Muhn, Valeri V. Nikolaev :: SSRN
Even without any narrative or industry-specific information, the LLM outperforms financial analysts in its ability to predict earnings changes. The LLM exhibits a relative advantage over human analysts in situations when the analysts tend to struggle. Furthermore, we find that the prediction accuracy of the LLM is on par with the performance of a narrowly trained state-of-the-art ML model.
10-Year Treasuries Are a ‘Screaming Buy’ as Fed Shows Credibility, BMO Says - Bloomberg
Lyngen sees a 3% yield on 10-year Treasuries sometime in 2024’s first half. “But we can easily close in a range of 3.5 to 3.75 this year,” he said. Real rates as measured by the spread between 10-year Treasuries and their inflation-protected peers should be in a range of 1.65% to 2.15%, “well within what the Fed wants to see,” Lyngen said.
Bond Bulls at JPMorgan, Allianz Keep Piling Into a Bet Gone Bad - Bloomberg
For his part, JPMorgan’s Michele is confident bond yields will fall once the Fed winds down its tightening cycle, long before the first rate cut.
“Whether the US economy enters recession or a soft landing, the bond market rallies after the last rate hike,” he said. “The Fed may keep rates at these levels for quite some time, but growth and inflationary pressure continue to slow.”
The chaos wrought by the FOMC keeps unfolding | National Mortgage News
"What's interesting is almost none [of my clients] have a lot of perspective about the mortgage business. A lot of them have never been in an environment that is predominantly purchase. I was raised in that environment. I spent my first 20 years in an environment where we didn't have refinances really. Everything was a purchase. I think they're struggling with that. And then the second shoe that hit them all is margin compression, and product compression, which they weren't expecting. Almost everybody's business model that I saw underestimated the amount of compression on the gross revenue side of their business."
How a Disaster Expert Prepares for the Worst | The New Yorker
“We’re going to be using a lot of ‘when’ today. I don’t use ‘if.’ I don’t touch wood too much,” Easthope said. “It is when, when these emergencies happen.” In spite of her experiences witnessing protocols fail or be subverted, Easthope still argues strongly for disaster and recovery plans. She has no time for people (normally male uniformed commanders) who believe they are dealing with something that no one has ever suffered through before. “Don’t go out there with ‘unique.’ Don’t go out there with ‘unprecedented.’ Don’t go out there with ‘This was a terrible incident we couldn’t have foreseen,’ ” she said. “You are walking plowed, furrowed fields that other people have walked, and they are there to guide and support you.”
Can the US Dodge a Recession? This Economist Thinks So - Bloomberg
“In science we use models all the time, and they’re simplifications of reality,” he said. “And part of the skill of the scientist is to know when to deploy the model and when not to or, in other words, to know the limitations of the model. And maybe I’m in a good position of knowing the limitations, given that it’s my model.”
Top performers don’t always provide the best advice – Research Digest
The analysis showed that the best performers believed that they had given the best advice. However, when pieces of advice were then given to a fresh group of players, this turned out not to be the case: players who were given advice from the best performers didn’t improve at the game any more than players given advice from from other performers.
The researchers also report a fascinating extra finding from this study: the second group of players rated advice that had come from the top performers as being the best — even though they had no knowledge of these people’s performance.
In a fresh study, the team considered some possible explanations for this. Perhaps the top performers gave more articulate or more authoritative advice, for example. In fact, they found that it was the number of independent suggestions that mattered — and top performers offered more. “In short, advice from the best performers was not better. It just sounded better because there was more of it.”
James Meek · What are you willing to do? On the case for civil war · LRB 26 May 2022
Systemic Peace hasn’t publicly updated its ratings for most countries since 2018, but for reference, the eight countries rated 5 that year were Ecuador, Haiti, Mali, Mozambique, Niger, Papua New Guinea, Somalia and Suriname. The ways Systemic Peace’s data diverge from what a lay person would expect of a democracy-autocracy scale are interesting. The US is rated as a sound democracy from 1829 until just before the Civil War, despite its embrace of slavery in that period. Belgium scores a solid 6 for much of its brutal rule over Congo. The UK gets a perfect 10 rating from 1922, despite being, at that time, at the head of a racially organised, exploitative empire that denied democratic rights to millions. Walter claims that a polity score is the best predictor of a country’s instability, but Systemic Peace gives Britain a 10 throughout the period of the Troubles in Northern Ireland, which Walter uses elsewhere as an example of an actual civil war.
Wall Street’s Lack of Big Tech Skepticism Is Why It Got Meta Earnings So Wrong - Bloomberg
While analysts who advised clients to buy Meta ahead of its 26% plunge could be in for some tough conversations, data compiled by Bloomberg show that the two in Europe who recommended selling shares have been saying so since 2015 and 2019, respectively—so the calls may not have been that useful to their clients. Even after the post-earnings plunge, shares are up about 190% since the beginning of 2015.
Alzheimer’s: Inflammatory markers are conspicuous at an early stage: Evidence of damage and also neuroprotective processes long before symptoms of dementia manifest -- ScienceDaily
n recent years, it has become evident that the brain's immune system and related inflammatory processes -- also known as "neuroinflammation" -- significantly contribute to the development of Alzheimer's disease. In view of this, the scientists analyzed various immunological biomarkers that are characterized by good detectability in the cerebrospinal fluid and reproducible results. "It was already known that these markers indicate immune processes in the context of Alzheimer's disease. However, how these markers relate to brain volume, cognitive performance and other parameters had not been studied as comprehensively as we have now," explains Prof. Michael Heneka, who led the current study during his long-time tenure at DZNE and UKB. Since the beginning of this year, he has been director of the Luxembourg Centre for Systems Biomedicine.
"We have found that some of these inflammatory markers are conspicuous even when there are no symptoms of dementia yet," Heneka says. "Based on the data we have so far, we can't specify the lead time at this point. But my estimate is that it is at least ten to twenty years."
Differences in financial risk preferences can make or break a marriage: Couples who disagree on savings and investment decisions are twice as likely to divorce -- ScienceDaily
After controlling for an array of household characteristics, such as the wife's and husband's education level, religion, region of origin, cultural background and more, the author found differences in risk preferences are the biggest predictor of marital separation in the long run.
Couples who had the most dissimilar risk attitudes are twice as likely to divorce, compared to couples with the most similar preferences. Additionally, of all the risk categories listed in the survey, differences in financial risks were the strongest predictor of divorce.
Focus on outliers creates flawed snap judgments: Our quick scan of a crowd isn't as reliable as we think, new research suggests -- ScienceDaily
For the study, which appears online in the journal Cognition, researchers recruited 48 observers ages 18-28. Participants were presented with a grid of 12 faces and were given just one second to glance at the grid. Study participants were then asked to estimate the number of men and women in the grid.
Participants accurately assessed homogenous groups -- groups containing all men or all women. But if a group contained fewer women, say, participants overestimated the number of women present.
The researchers also tracked participants' eye movements. They found that participants looked more often at whichever group was in the minority -- men or women.
All of this occurred very quickly -- during a glance of just one second, said co-author and Duke psychologist Scott Huettel.
"We should recognize that our visual system is set up to orient ourselves towards some types of information more than others," Huettel said. "People form an initial impression very quickly, and that impression biases where we look next."
Interestingly, the same tendency to focus on the outlier also extended to scanning other kinds of images.
In a second experiment, study participants were shown a grid of nature photos showing a variety of indoor and outdoor scenes. Participants consistently overestimated whatever type of scene appeared less often.
For instance, if a grid of 12 photos contained two outdoor scenes -- say, a waterfall and a mountain range -- participants reported, on average, that the grid contained three such scenes.
Damage to white matter is linked to worse cognitive outcomes after brain injury -- ScienceDaily
The most unexpected aspect of our findings was that damage to gray matter hubs of the brain that are really interconnected with other regions didn't really tell us much about how poorly people would do on cognitive tests after brain damage. On the other hand, people with damage to the densest white matter connections did much worse on those tests," explains Justin Reber, PhD, a UI postdoctoral research fellow in psychology and first author on the study. "This is important because both scientists and clinicians often focus almost exclusively on the role of gray matter. This study is a reminder that connections between brain regions might matter just as much as those regions themselves, if not more so."
Researchers develop a method for predicting unprecedented events -- ScienceDaily
Bray settled on three eclectic datasets: an eight-year study of plankton from the Baltic Sea with species levels measured twice weekly; net carbon measurements from a deciduous broadleaf forest at Harvard University, gathered every 30 minutes since 1991; and measurements of barnacles, algae and mussels on the coast of New Zealand, taken monthly for over 20 years.
The researchers then analyzed these three datasets using theory about avalanches -- physical fluctuations that, like black swan events, exhibit short-term, sudden, extreme behavior. At its core, this theory attempts to explain the physics of systems like avalanches, earthquakes, fire embers, or even crumpling candy wrappers, which all respond to external forces with discrete events of various magnitudes or sizes -- a phenomenon scientists call "crackling noise."
Built on the analysis, the researchers developed a method for predicting black swan events, one that is designed to be flexible across species and timespans, and able to work with data that are far less detailed and more complex than those used to develop it.
"Existing methods rely on what we have seen to predict what might happen in the future, and that's why they tend to miss black swan events," said Wang. "But Sam's method is different in that it assumes we are only seeing part of the world. It extrapolates a little about what we're missing, and it turns out that helps tremendously in terms of prediction."
Why Doctors Are Bad At Stats — And How That Could Affect Your Health
Gerd and his team have explored whether medical professionals understand the statistics measures actually needed to prove that a cancer screening programme saves lives.
This is a classic problem in health statistics. What clinicians need to compare is mortality rates, not 5-year survival rates. The mortality rate tells the number of deaths in a period of time. In contrast, the 5-year survival rate only tells how many people live 5 years after the day they have been diagnosed with cancer.
Some screening programmes can diagnose people earlier — which can increase those ‘5-year survival rates’ — without making them live any longer.
How expectations influence learning -- ScienceDaily
Using functional magnetic resonance images, the researchers were able to show that different brain connections between the prefrontal cortex and the thalamus were responsible for maintaining a learning strategy or changing the strategy. The higher the expectations before the decision, the sooner the strategy was maintained and the lower the strength of these connections. With low expectations, there was a change of strategy and the regions seemed to interact much more strongly with each other. "The brain appears to be particularly active when a learning strategy has to be changed while it takes significantly less energy to maintain a strategy," concludes Pleger.
Coronavirus: DeWine criticizes those defying orders, including churches; cases reach 2,547 as Ohio designates hospital zones - News - The Daily Record - Wooster, OH
Coronavirus continues its relentless rise in Ohio — advancing toward a projected peak later this month — as 348 more infected Ohioans and 10 more deaths were reported Wednesday.
In his daily briefing, Gov. Mike DeWine hinted that extended or new restrictions on residents and businesses were coming as soon as Thursday. He also outlined efforts to expand hospital capacity, including use of convention centers for patients without COVID-19.
With limited testing concealing the true extent of the pandemic, total cases rose 16% to at least 2,547 and deaths increased by 18% to 65 total since the virus first was confirmed in Ohio on March 9, state figures show.
The Mathematics of Predicting the Course of the Coronavirus | WIRED
According to IHME’s models, 41 states will need more hospital beds than they currently have. Twelve states will need to boost their numbers of ICU beds by 50 percent or more. The models predict that over the next four months, these shortfalls will contribute to the deaths of 81,000 Americans, with the number of deaths per day peaking as soon as mid-April.
The Eerie Parallels Between Coronavirus and the Bubonic Plague - Rolling Stone
Crises like these — whether it’s a crisis of political legitimacy, or a pandemic that demands response, or some kind of major external war that crops up out of nowhere — the chances are good that whatever snaps under the pressure of that crisis was probably straining already, was probably barely chugging along already. There’s some kind of deep problem that a crisis is going to expose, bring to the fore, and then break very dramatically for everybody to see.
We see the crisis and we see the break — and we equate the two. We’re narrative creatures. That’s how we understand the world. We understand things as a story with a climax, and the break has to be the climax. It’s very hard for us to turn a more analytical eye and see the collection of very small things that lead up to a systemic break. It’s just difficult. But these disasters don’t create these trends so much as they supercharge them.
More pablum from experts
A global, novel virus that keeps us contained in our homes—maybe for months—is already reorienting our relationship to government, to the outside world, even to each other. Some changes these experts expect to see in the coming months or years might feel unfamiliar or unsettling: Will nations stay closed? Will touch become taboo? What will become of restaurants?
Pandemic dodged, but black swan looms
Billionaire investor Hamish Douglass says he is confident the coronavirus outbreak will not develop into a global pandemic that will roil the global economy, but has warned that the probability of a black swan event such as a cyber attack, terrorist attack or pandemic is increasing.
The founder of Magellan Financial Group, who spoke to AFR Weekend before he took the stage in front of 2500 Magellan investors at an event in Melbourne, said he had been impressed with the Chinese government’s response to the crisis.
“Could you imagine if something like this actually broke out in the United States?” Douglass said.
Pandemic dodged, but black swan looms
Billionaire investor Hamish Douglass says he is confident the coronavirus outbreak will not develop into a global pandemic that will roil the global economy, but has warned that the probability of a black swan event such as a cyber attack, terrorist attack or pandemic is increasing.
The founder of Magellan Financial Group, who spoke to AFR Weekend before he took the stage in front of 2500 Magellan investors at an event in Melbourne, said he had been impressed with the Chinese government’s response to the crisis.
“Could you imagine if something like this actually broke out in the United States?” Douglass said.
Fragmented physical activity linked to greater mortality risk -- ScienceDaily
The researchers found that for each 10 percent higher activity fragmentation there was a 49 percent increase in the risk of mortality. The researchers defined activity fragmentation as the probability of transitioning from an active state to a sedentary state for each participant, so shorter average activity periods meant higher fragmentation.
The researchers also analyzed the duration of each participant's bouts of activity, and found that "percent of activity spent in bouts of less than five minutes" appeared to be another good marker of mortality risk. Each additional 10 percent of active time spent in such short bouts was associated with a 28 percent increase in the chance of mortality. Percent of active time spent in 5- to 10-minute bouts was not a significant indicator of mortality risk.
