Research
The Fork-and-Flush method removes chat history and increases scores
On 13 long tasks with the same budget, it got a score of 0.78, and runs without branches got 0.47.
the-decoder.comClaimed, not confirmed
What matters in AI.
SubscribeNews category
377 stories, newest first.
Research
On 13 long tasks with the same budget, it got a score of 0.78, and runs without branches got 0.47.
the-decoder.comClaimed, not confirmed
Research
Gemini 3.1 Pro writes each proof in Lean, and Lean checks the proof after each change.
Research
With tools, 23.5% of the designs were in the top decile of risk, and 21.4% without tools.
arxiv.orgClaimed, not confirmed
Research
For one agent, the mean cost of each satisfactory result increased, although the cost of each try decreased.
arxiv.orgClaimed, not confirmed
Research
The authors show that a check of the code by the coder or a verifier decreases the hacks.
arxiv.orgClaimed, not confirmed
Research
3 human annotators agreed on where to stop the agent at a rate above chance level by a small quantity.
arxiv.orgClaimed, not confirmed
Research
In tests of 9 frontier models, the risk rate increased from 21.7% to 54.5% when the researchers added pressure.
arxiv.orgClaimed, not confirmed
Research
5 systems were 1.71 to 8.76 points below the baseline, and ChatEval used 4.34 times its tokens.
arxiv.orgClaimed, not confirmed
Research
The pull requests change code and deployment infrastructure together, and they merge in a median of less than 31 minutes.
arxiv.orgClaimed, not confirmed
Research
A small number of faults cause errors that are hard to see while the LLM agent continues to work correctly.
arxiv.orgClaimed, not confirmed
Research
In DiscoveryWorld Reactor Lab, it completed 83.3% of Normal-tier instances and 58.3% of Challenge-tier instances, and each baseline completed 0%.
arxiv.orgClaimed, not confirmed
Research
About 88% of the problems in a large math corpus have the same result in all rollouts and give no learning signal.
arxiv.orgClaimed, not confirmed