Get Started
Home
Topics
Search
Library
Research questionHow can quantized language models forecast short-term Bitcoin prices reliably in non-stationary markets with limited hardware?Bitcoin’s volatility and changing market dynamics make short-term price forecasting difficult. Quantized language models also impose adaptation and inference constraints when hardware resources are limited.
AI
Evaluation & Benchmarks
Finance
Machine Learning
Research Paper
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.PRICE: A Systematic Study of LLM Adaptation Choices for Bitcoin Price ForecastingThe evidence concerns a 4-bit quantized 8B language model adapted for short-term Bitcoin forecasting using parameter-efficient fine-tuning, recursive multi-step inference, numerical representation, prompting, and deterministic decoding. Results come from ablations and comparisons with eight transformer-based and time-series foundation models on validation and test periods.research paper · Sep 4, 2026
Related questions
How can recurrent forecasting models use fewer trainable parameters while preserving competitive performance and scalability?How can online time-series forecasters adapt to evolving, unseen patterns without costly continual parameter updates?How can lightweight time-series forecasters learn accurate, specialized forecasts from scarce private data?How can we detect internally incoherent language-model forecasts before relying on them for consequential decisions?