Hugging Face daily papers·3d agoIterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis#llm-agents#deep-search#reinforcement-learning
arXiv cs.AI / cs.LG / cs.CL·4d agoGrow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents#llm-agents#agent-harness#webarenaAI research
The Decoder·13d agoIris-mini and Iris-pro are the strongest open-weight search agents in their class#allspark#browsecomp#iris-mini 5 min1
MarkTechPost·15d agoCan LLMs Engineer Their Own Agent Harness? ByteDance Seed’s HarnessDev Says Only 34 of 64 Changes Generalize#agent-harnesses#browsecomp#bytedance-seed 4 min1
