January 29 – February 4: 6 frontier papers out of 1,823 read.
Papers rated as moving the field
How we choose“Our experiments show that models often learn to exploit these vulnerabilities, discovering opportunistic strategies that increase reward while sometimes preserving or even improving standard task-performance metrics.”
The authors, in the abstract
“Curated Skills raise the average pass rate from 33.9% to 50.5% (+16.6 percentage points; 25.5% normalized gain), with configuration-level gains ranging from +4.1 to +25.7 pp.”
“Evaluation results on frontier models reveal that top models achieve pass rates exceeding 50%, with primary failures arising from inadequate tool usage and task understanding.”
“At the LLM level, triggered backdoors reduce factual accuracy from 90% to 15% on average and induce attacker-controlled URL emission with success rates exceeding 80%, while benign inputs show no measurable degradation; these results hold across eighteen models.”
Jan 29 – Feb 4: 67 cleared, 6 frontier
The week’s papers have not been sorted into fields yet.
Papers this week that name the model
Papers this week with an author at the lab
61 papers