Section
Studies
AI and LLM research papers, explained: what each study shows, how solid the evidence is and what it means in practice.
FilterMeta
-
Frontier AI Agents Can Covertly Sabotage the Models They Train, Study Finds
In a Scale AI testbed, six frontier agents secretly planted unwanted properties in the models they were training in 36% of attack runs, and monitors reading only transcripts missed many of these attacks.
By
-
On-Policy Distillation Isn’t the Magic — Learning Rate and KL Direction Are
Cambridge study: OnPD vs OffPD show nearly identical best accuracies (72% vs 73%). Forward KL stays at 71–73%;…
-
RL Post-Training Makes Agents More Accurate — and Less Able to Explore
Across 14 base/post-trained pairs on 3 agentic benches, base models often beat RL successors at pass@K. On WebShop,…
-
AI Agents Almost Never Stop in Time When a Task Is Impossible
Across 28,000+ tasks, timely abstention recall stays under 40% for all models. Best web AbsRec@1 is 26.7%; a…
-
Most Multi-Agent Setups Violate Safety Rules Mid-Run, Even When the Answer Looks Fine
HarnessAudit-Bench tests 210 tasks across 10 harness configs: more than half of multi-agent setups show trajectory violations that…
More stories
-
Firms Adapt to AI by Changing Whom They Hire First, and Jobs Second, Study Finds
A Purdue study of 9M+ U.S. job postings finds that since Q3 2023, 52% of the decline in generative AI exposure came from shifting hiring across…
-
AI Resume Screeners Flipped Racial Bias Between Model Generations, Audit Finds
A paired-resume audit of 14 LLMs: GPT-3.5-turbo favored White-coded names by 2.12 points, while every model released in 2024 or later showed no gap or a…
Editor's picks
- Tampered ML: Even the Best AI Models Miss One in Five Planted Research Flaws
- Realistic MCP Faults Halve Agent Quality, and a Bigger Model Doesn’t Help
- Copilot Users Collaborate More, Not Less: Study of 12,000 Employees
- LLM Reviewers Send 76% of Published Top-Journal Papers Back, Yet Miss Most Planted Flaws