Agents
2 agents that work in turn get a medal on 78.2% of MLE-bench tasks
The method is 8.0 points above the top open-source baseline and uses a quarter of its time.
arxiv.orgClaimed, not confirmed
News category
Agents
230 stories, newest first.
Agents
RATED cut the fault selection rate of agent tools from 86.2% to 2.9%
In a test, 30% of API calls did not work, and the added time was less than 1 millisecond.
arxiv.orgClaimed, not confirmed
Agents
ARIA selects when a low cost model must get help from a stronger model
ARIA recovers 44% to 81% of the success gap at less than 50% of the cost of the advisor model.
arxiv.orgClaimed, not confirmed
Agents
ARTIS lets an AI agent test its actions in a simulation first
The authors say that the simulator must be accurate for rare failure modes that have a large effect.
arxiv.orgClaimed, not confirmed
Agents
CLEAT makes an AI agent select when to get a clarification
With Qwen3-4B, CLEAT has the highest mean score in all 10 tests on 4 benchmarks.
arxiv.orgClaimed, not confirmed
Agents
TerminalHorizon makes 1.5K task environments for terminal agents
The authors say that Qwen3.5-27B increases its results on all 12 benchmarks after the tuning on this data.
arxiv.orgClaimed, not confirmed
Agents
Google adds the Gemini Agent to Gemini Enterprise
The agent takes a goal, makes a plan, and picks the AI model, which can be Claude from Anthropic.
Agents
Sakana AI shows MASS, a method that trains a model on agent teams
MASS uses one model as executor, optimizer and evaluator, and it gives more score for each output token than the base model on research benchmarks.
pub.sakana.aiClaimed, not confirmed
Agents
Ankur Duggal found bad output in 44% of the runs of an agent
The traces of the agent showed the fault, because they show each tool call and each document that it used.
Agents
2 AWS engineers say hooks stop an AI agent with a fault
Hooks are deterministic Python code in the agent loop, and the engineers do not try to control the agent from the external side.
Agents
Olyzon agents move live ad budgets across 4 platforms
Olyzon says a person sets the budget limits and the agents work in those limits.
Agents
Ant Group, ByteDance and Tencent make personal AI agents
Ant Group wants to release LingGuang 2.0, an agent with long-term memory, this month.