Re-Search tests whether LLMs can rediscover published research ideas
On 200 ICML 2026 papers, six LLMs make nearly all progress in the first 30 of 100 yes-or-no questions.
Stories per day · 30 days
Relatedllm-evaluation2benchmark1benchmarks1mechanistic-interpretability1research-ideation1scientific-agents1scientific-judgment1weight-editing1
9 stories, newest first · one card per story
Stories tagged #autoresearch
LLMs can read and write neural network weights
Research shows LLMs can directly read and edit neural network weights, enabling autonomous model improvement that cuts traditional autoresearch out of the loop.
1
SciΠvot: frontier models lag humans on research judgment
Best project-condition accuracy is 63.5%, versus 73.4% for fresh researchers on the same cases.
Clearing chat history helps AI agents find better solutions
Microsoft Research shows long-running AI agents plateau in "idea basins"; a Fork-and-Flush context-reset method boosts scores 66% on equal compute.
8
Optimizing the Optimizer: Language Models Discover Faster Molecular Relaxation
An agent rewrote Sella into AutoSella, needing only 40.2–77.2% of Sella's force calls at r2SCAN-3c.
AutoMark: Enabling Autoresearch to Discover Better LLM Watermarks
AutoMark uses frontier-model agents to autonomously discover distortion-free LLM watermarks that beat prior schemes.
AutoResearch at Production Scale: Failure Modes and a Multi-Agent Framework
Production AutoResearch on book embeddings hit five failure modes; a scaffolding fix lifted Recall@6 1.82x.
Agora: Git as Shared Memory for Collective AutoResearch
Agora records multi-agent research as an append-only Git DAG; 13 LLM workers ran nearly 12 days on a weight-transfer problem.
[AINews] 10% worse, 100x cheaper, 10000x faster: Why Simulation is taking over
Latent Space argues AI training pipeline stages—rewards, data, teachers, curricula, environments—are flipping from human-made to model-made simulation.
7