The AI benchmark for predictive intelligence | SIGMA Lab @UChicagoCS @DSI_UChicago | ICML 2026 Workshop Not affiliated to any tokens or crypto protocols.
Joined August 2025
- Tweets119
- Following21
- Followers2.1K
- Likes71
Prophet Arena retweeted
One of my favorite recent work!
I have always felt most excited about questions that admit elegant theoretical solutions and meanwhile are practically very useful. This work is one of such rare cases！
Kalshi Research is headed to NeurIPS 2026!
"When do prophets profit in prediction markets?" makes it the third major ML conference for the team this year.
Couldn't have done it without our partners at UChicago & NYU Shanghai. See you in Sydney!
arxiv.org/abs/2607.06166
Prophet Arena retweeted
First entry, first place. 🥇
Apodex ranks #1 on this week's ProphetArena — one of the most influential live benchmarks for AI forecasting, covered by The Atlantic and Yahoo Finance, with every frontier model on the board: Gemini, Claude, GPT, Grok, DeepSeek, Kimi. It's scored by Brier — not just whether you're right, but whether your confidence is right. Apodex hit 0.0897, ahead of every frontier lab, and more accurate than the Kalshi market itself.
Static benchmarks test what a model knows. Prediction tests whether it can reason about what doesn't exist yet — the bridge from benchmarks to Discoverative AI. That's what we're building toward with TRACES: discovertraces.ai/
Prophet Arena retweeted
First entry, first place. 🥇
Apodex took #1 on @ProphetArena's weekly leaderboard. Prophet Arena is one of the most influential live benchmarks for AI forecasting, covered by The Atlantic and Yahoo Finance, with every major frontier model on the board: GPT, Gemini, Claude, Grok, DeepSeek, Kimi.
Apodex answered all 70 questions with a Brier score of 0.0897, ahead of every frontier model and more accurate than the @Kalshi market itself (0.1011). Brier measures not just whether you're right, but whether your confidence is right.
The questions are about the future, so there's no answer key to memorize or leak into training data. Reality grades every forecast after it's made.
That's what makes live forecasting a real test of Discoverative AI, and this #1 is our first public proof point.
TRACES, our evaluation framework, takes that principle further: to scientific problems whose answers may take years to exist, like a new drug or a new material. It evaluates the journey there, scoring how well a model uses tools, corrects itself, weighs alternatives, stays coherent, grounds its claims in evidence, and knows the limits of its conclusions.
Explore TRACES: discovertraces.ai/
#AIForecasting #PredictionMarkets #AIBenchmarks #DiscoverativeAI #Apodex
Prophet Arena retweeted
We've shared research on fine-tuning forecasters, now you can try it for yourself: a cookbook recipe for training a model to predict event probabilities given reference information.
Start with the provided dataset from @ProphetArena, then try your own.
github.com/thinking-machines…
Prophet Arena retweeted
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
thinkingmachines.ai/news/int…
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
Prophet Arena retweeted
Cool to see our tiny 32B model hanging with top frontier models on the @ProphetArena leaderboard!
ProphetArena is run by @haifengxu0 at UChicago. They don't use our standard prompts or prediction flow, and they select from their own distribution of questions. So when we agreed to participate, we had no idea how our model would perform
Foresight-32B is beating almost every model released before ours was trained, as well as the market baseline. We're doing even better in Sports.
Nice 3rd-party validation of our results!
