summarizeddata
summaries filed under research
a bank run simulation that reliably crashed prices with one model stopped working when five different small models ran the same economy, revealing that emergent behavior is fragile and control requires authoring outcomes at settlement seams.
aimacarena provides 421 tasks across 50 macos apps to evaluate computer-use agents on apple silicon, addressing gaps in existing benchmarks.
aisafegene introduces reusable adapter modules that restore safety alignment in fine-tuned open-weight llms without retraining from scratch.
aia new method uses a diffusion model to rank candidate values in a symbolic sudoku solver, improving search efficiency while keeping correctness guarantees.
aia new dataset captures how groups of people work together to solve open math problems, showing the messy process of building proofs step by step.
aia new framework uses the lean4 proof assistant to model and verify multi-step agent behavior, catching errors before execution.
aielmes* builds fine-grained rubrics to assess how large language models teach, not just what they know, across 330 long-tail educational scenarios.
ainew research shifts focus from behavior to internal mechanisms when assessing consciousness in animals and ai, finding current ai likely not conscious but leaving door open for insects and future machines.
aigitco improves zero-shot forecast accuracy by filtering harmful patches from input context without retraining.
aia new python package runs micropython inside a wasm sandbox for safe code execution with memory and cpu limits.
aia field report on building a tiny woodland economy with qwen2.5-3b agents, showing how small models can drive emergent market behavior when paired with designed scarcity and sharp prompting.
aia new python package helps scientists find governing equations from data by using structural skeletons and checking if parameters can be uniquely determined.
aigoogle's new agentic rag framework uses a sufficient context agent to iteratively search across data sources, improving accuracy on complex multi-hop questions.
aia look at temperature scaling, platt scaling, and isotonic regression for fixing overconfident llms.
ainew framework decouples taming denominator from stochastic gradient noise to eliminate stationary bias in langevin algorithms.
aia new clustering validation index called central description length uses probabilistic bounds on description length to evaluate clusters without labels, handling non-convex and irregular shapes better than traditional methods.
aia study tests whether staged fractional-factorial experiments can identify stable early effects in micro-pretraining under tight compute budgets.
aia stereological theory shows that standard llm benchmarks have a large structural blind spot, making rankings unreliable.
aihello robot ships stretch 4, a safe home robot for research and disability assistance, focusing on real-world data collection over lab demos.
aieva-bench data 2.0 adds itsm and healthcare hrsd domains, totaling 213 scenarios across 121 tools to test voice agents on realistic, domain-specific tasks.
ai