SRA-Bench
PulseAugur coverage of SRA-Bench — every cluster mentioning SRA-Bench across labs, papers, and developer communities, ranked by signal.
-
New 'Capability Pages' method boosts LLM skill retrieval accuracy
Researchers have developed a new method called Capability Pages to improve how large language model agents retrieve reusable skills. This approach formalizes a skill's capability as its executable region, the set of que…
-
New SkillSight framework boosts LLM agent skill retrieval accuracy
Researchers have developed SkillSight, a novel framework designed to improve the accuracy and efficiency of skill retrieval for large language model agents. SkillSight addresses the issue of shared descriptive patterns …
-
New Skill Retrieval Augmentation paradigm boosts agentic AI performance
Researchers have introduced Skill Retrieval Augmentation (SRA), a new method for enhancing agentic AI systems. SRA allows agents to dynamically retrieve and apply skills from large external corpora, overcoming the limit…