Researchers compared standard fine-tuning with a different method called Implicit Chain of Thought (ICoT). They looked at how models handle long calculations and whether the models can keep intermediate results during many steps.
Under standard fine-tuning, models with two to 12 layers had less than 1% accuracy on four-digit multiplication. By contrast, the ICoT-trained model reached 100% accuracy. The team found they could read running sums from the ICoT model’s internal states, which shows the model learned to remember useful intermediate values.
The researchers also added a training objective that teaches a model to track running sums. Adding this objective raised a two-layer model’s accuracy to 99% without explicit chain-of-thought supervision.
Difficult words
- fine-tuning — small training to improve a model
- implicit — not directly shown or written
- intermediate — between the first and last steps
- accuracy — how often answers are correct
- internal — inside a system or model
- objective — a goal used during training
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Do you write intermediate results when you do long calculations? Why?
- Would you like a model that remembers intermediate values? Why or why not?
- How can remembering intermediate values help solve a problem?
Related articles
New AI tools for tuberculosis shown at lung health conference
Researchers presented four new AI approaches for detecting and monitoring TB at the Union World Conference on Lung Health in Copenhagen (18–21 November). The tools include breath analysis, cough screening, vulnerability mapping and a chest X‑ray tool for children.
Indonesia tightens rules for digital platforms
Indonesia is increasing regulation of global digital platforms to curb misinformation and protect public safety. Officials inspected a major company's office, require platform registration, and use takedown systems, which has drawn criticism over unclear rules and rights.
Small pause to slow misinformation on social media
Researchers at the University of Copenhagen propose a small pause before sharing on platforms like X, Bluesky and Mastodon. A computer model shows that a short delay plus a brief learning step can reduce reshares and improve shared content quality.
Electric car batteries can power homes and cut costs
A University of Michigan study finds that using electric vehicle batteries to power homes (vehicle-to-home, V2H) can save owners thousands of dollars and reduce greenhouse gas emissions. Results differ across regions and the technology is still being tested.