MLE-World: Predicting Machine Learning Experiment Outcomes with a Learned World Model
MLE-World is a world model finetuned from Qwen3.5-4B that predicts a machine learning program's terminal output, including error traces, training logs, and validation scores, from its code and data description without executing it, along with a confidence score. Because each prediction costs one inference call rather than a training run, it makes test-time scaling affordable for AI research agents. Five test-time scaling policies guided by the model compare candidates by predicted outputs, and candidates are executed for real only when confidence is low. On 17 MLE-bench Lite tasks, it matches real terminal output on 77.9% of held-out programs versus 41.1% for much larger prompted models.
Open original source huggingface.coOne primary source; other links are context, not independent corroboration.
- Primary sourceHugging Face Daily Papers
- Discovery linkHugging Face
- Linked in sourceView arXiv page
- Linked in sourceView PDF
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Open full article view →- Original paper+45
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
MLE-World is a world model finetuned from Qwen3.5-4B that predicts a machine learning program's terminal output, including error traces, training logs, and validation scores, from its code and data description without executing it, along with a confidence score. Because each prediction costs one inference call rather than a training run, it makes test-time scaling affordable for AI research agents. Five test-time scaling policies guided by the model compare candidates by predicted outputs, and candidates are executed for real only when confidence is low. On 17 MLE-bench Lite tasks, it matches real terminal output on 77.9% of held-out programs versus 41.1% for much larger prompted models.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: huggingface.co. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Results are reported on 17 MLE-bench Lite tasks only; source claims not independently checked, and predictions can be wrong.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- MLE-World predicts program terminal output from code and data description without executing the program.
- It matches real terminal output on 77.9% of held-out programs, versus 41.1% for much larger prompted models.
- MLE-World guided test-time scaling achieves up to 22.1% higher observed mean performance than the agent without a world model.
- Candidate programs are run for real only when the prediction's confidence score is low, reducing real execution.
Results are reported on 17 MLE-bench Lite tasks only; source claims not independently checked, and predictions can be wrong.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
Executing candidate ML programs is expensive, which limits how many options an AI research agent can explore. A learned predictor of experiment outcomes could make broader search affordable, with real execution reserved for uncertain cases.
Limits and open questions
Results are reported on 17 MLE-bench Lite tasks only; source claims not independently checked, and predictions can be wrong.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.