aiqbenchInformații despre AI & calcul cuantic

...

Română · beta. Traducerea a fost făcută automat și este încă în curs de revizuire. Citește pagina în engleză

perspectiva aiqbenchCercetare AI

TechPulse take: Google Research has released RRSI, a framework enabling large language model agents to iteratively refine their own prompts and tools while keeping model weights frozen. By incorporating specific constraints like leakage critics and cost rules, the system prevents overfitting to benchmark tasks and ensures improvements generalize to unseen data. This approach offers a robust method for autonomous agent self-improvement without requiring retraining of base models.

Sursă care acoperăMarkTechPost

Ce este aceastaLectura unui model asupra articolului și a acoperirii conexe, scrisă pentru aiqbench. Nu sunt cuvintele publicației.

Google Research Open-Sources RRSI: AI Agents That Improve Their Own Harness Without Overfitting

MarkTechPost29 septembrie 20265 min de cititArticolul și perspectiva sunt în engleză.

Google Cloud AI Research has open-sourced RRSI, a framework that lets LLM agents rewrite their own prompts, tools and memory while model weights stay frozen. It adds a leakage critic, a noise floor, a cost rule and pruning so gains carry over to new tasks. With Claude Opus 4.8, Terminal-Bench 2.1 rose from 74.2% to 80.2%, and all 6 held-out splits improved. The post Google Research Open-Sources RRSI: AI Agents That Improve Their Own Harness Without Overfitting appeared first on MarkTechPost.

Google Cloud AI Research, in collaboration with UNC-Chapel Hill, Stanford, and Washington University in St. Louis, has released RRSI (Regularized Recursive Self-Improvement). It enables an LLM agent to rewrite its own harness: prompts, tools, memory, control flow, and sub-agents. Model weights remain unchanged. RRSI constrains the improvement loop itself, ensuring that gains hold on benchmarks the agent never optimized against. Deployable? Yes, as a research framework. The code is Apache 2.0, requires Python 3.10+, and accepts any LiteLLM model string. Defaults assume Claude Opus 4.8 on…

Extras afișat conform termenilor RSS ai publicației · 86 cuvinte · textul integral și figurile sunt pe site-ul publicației

Citește pe MarkTechPost ↗Întreabă despre acest articol