FrontierPhysics evaluates AI agents on end-to-end physics research
FrontierPhysics tests 22 models across 56 physicist-authored tasks and finds no configuration publishes more than 46 of 168 rollouts.
Stories per day · 30 days
2 RSS
Relatedai-agents1autonomous-research1benchmark1benchmarks1coding-agents1end-to-end-ai1experiment-evaluation1frontier-models1hypothesis-generation1life-science1mathematical-bounds1physics-benchmark1scientific-discovery1
10 stories, newest first · one card per story
Stories tagged #research-agents
Frontier AI agents propose experiments with limited ranking gains
Six frontier models propose MVEs from NLP papers but show minimal improvement in experiment selection over random picks.
Épi agent reports 120 record-setting autonomous research results
Authors say Épi-turbo beat Codex by 5.4% on 30 FrontierOR tasks at 64% of the model cost.
DiscoveryRSI recursively learns to generate research hypotheses
Automated scores rose from 4.86 to 5.39 after two self-improvement rounds across four life-science domains.
A Society of Researchers: Designing Institutions for Populations of Autonomous Research Agents
Paper proposes explicit institutions for populations of thousands of autonomous research agents.
DAGent: Evaluate-then-Grow Planning for Deep Research Agents
DAGent plans deep-research agent graphs incrementally from evidence confidence, outperforming open baselines on three benchmarks.
Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts
A protocol finds natural research-agent explanations do not yet show confirmed predictive credit on Tox21 and OpenML.
Recursive self-improvement of AI research agents
AIDE^2 recursively rewrote itself for eight days and matched a strong human-engineered research agent.
ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents
ScienceBuddy released: interactive scientific agent workspace coupling harness evolution with model reinforcement learning for continual self-improvement across four scientific task families.
Scaling Automatic Research Agents via World Models
WMRL replaces environment execution with a world model in RL, accelerating research-agent post-training 3-4x and letting 4B/9B agents beat 48B/120B open-weight agents.