Collect and classify aging theories from scientific literature to untangle the web of causality in aging research.
What do all five-year-olds understand that no senior academic does? Aging. As universal as it gets, yet its mechanisms remain unclear.
For decades, stubborn researchers from different fields have shared hundreds of opinions on why we age. These opinions are called the Theories of Aging. Some are groundless and unconvincing, while others may reveal underlying causes of age-related decline.
Together, they form pieces of a single puzzle – one whose solution could prevent more than 100,000 deaths every day. Let's solve it together!
The winner will be invited to write a paper together with ComputAge community.
In this challenge, you will collect and classify theories of aging. For example, you may find papers arguing that aging is driven by damage from free radicals, and others suggesting that the real cause lies in the diminished function of stem cells.
It may turn out that neither of these factors is central to aging. Or perhaps both represent independent mechanisms that contribute in parallel. Or could one even be the cause of the other?
The combinatorial complexity of such questions is huge, which is why we need to untangle the web of causality.
Gather as many theories of aging and corresponding papers as possible. Create a comprehensive database linking theories to scientific literature.
Use the collected papers to extract information about aging theories. For each paper, answer 9 research questions and write results in a table.
For each paper, answer these 9 critical questions:
Does it suggest an aging biomarker (measurable entity reflecting aging pace or health state, associated with mortality or age-related conditions)?
Yes, quantitatively shown / Yes, but not shown / No
Does it suggest a molecular mechanism of aging?
Yes / No
Does it suggest a longevity intervention to test?
Yes / No
Does it claim that aging cannot be reversed?
Yes / No
Does it suggest a biomarker that predicts maximal lifespan differences between species?
Yes / No
Does it explain why the naked mole rat can live 40+ years despite its small size?
Yes / No
Does it explain why birds live much longer than mammals on average?
Yes / No
Does it explain why large animals live longer than small ones?
Yes / No
Does it explain why calorie restriction increases the lifespan of vertebrates?
Yes / No
Your final score will be the average of your ranks in two parts:
EXAMPLE CALCULATION:
Rank #1 in Part 1 + Rank #3 in Part 2 = (1+3)/2 = 2.0 final score
Table 1: Aging theories
Columns: theory_id, theory_name, number_of_collected_papers
Table 2: Collected papers
Columns: theory_id, paper_url, paper_name, paper_year
Single CSV table with extracted answers
Columns: theory_id, paper_url, paper_name, paper_year, Q1, Q2, Q3, Q4, Q5, Q6, Q7, Q8, Q9
GROUND TRUTH TEST:
We have 10 hidden papers with known correct answers. Missing papers from Part 1 = incorrect predictions.
We expect participants to apply new LLM techniques to untangle the complex network of aging theories. While we believe this is a good task for a hackathon, we also recognize that it may be difficult to design a perfectly fair scoring system on the first attempt.
It is possible to game the evaluation process to achieve higher scores at the expense of producing a meaningful solution. We ask participants not to exploit the scoring system but to act in the spirit of the task.
You will be asked to share your code, and your solution will be evaluated holistically. The jury may adjust the scoring system during the competition if necessary.
Ready to map the aging landscape? Join the hackathon and help solve the puzzle of aging.