REVIEW 2 cited by
LLMs Can Teach Themselves to Better Predict the Future
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We present an outcome-driven fine-tuning framework that enhances the forecasting capabilities of large language models (LLMs) without relying on human-curated reasoning samples. Our method leverages model self-play to generate pairs of diverse reasoning trajectories and probabilistic forecasts for a set of diverse questions that resolve after the models' knowledge cutoff date. We then rank pairs of these reasoning traces by their distance to the actual outcomes before fine-tuning the model via Direct Preference Optimization (DPO). On a separate test set, our approach increases prediction accuracy of Phi-4 14B and DeepSeek-R1 14B by between 7--10\% over a base model and a DPO fine-tuned control model with randomized labels, bringing them on par with forecasting capabilities of much larger frontier models like GPT-4o.
Forward citations
Cited by 2 Pith papers
-
Prompt Engineering Large Language Models' Forecasting Capabilities
Across two preregistered studies and six LLMs, most prompt engineering variations produced no reliable improvement in forecasting accuracy, while Bayesian-style prompts consistently hurt performance.
-
Advancing Event Forecasting through Massive Training of Large Language Models: Challenges, Solutions, and Broader Impacts
A position paper advocating large-scale training of event forecasting LLMs, with proposals for label selection, counterfactual training data, auxiliary rewards, and multi-source datasets.
Discussion (0). Continue with ORCID to comment.