A new arXiv paper (2609.25237v1) proposes a post-training delivery benchmark that tests whether LLM agents can be trusted as forward-deployed engineers (FDEs), rather than whether they can raise a met…
Researchers have developed object-agnostic grasp planners that synthesize contacts directly on the gripper's geometry, enabling zero-shot grasping of unseen objects. GOAG achieves an 86.93% average su…
Microsoft researchers introduced LoRA (Low-Rank Adaptation) in 2021, a parameter-efficient fine-tuning technique that reduces trainable parameters by approximately 10,000 times and GPU memory requirem…
Deep Recurse released Ornith 1.5 35B-A3B, a mixture-of-experts language model with 35 billion total parameters and 3 billion active per token, built on a Qwen3.5-MoE architecture for coding, reasoning…
Ornith 1.5 9B, a 9-billion-parameter dense language model from ornith-ai, failed real-world agentic and coding tasks in a local test on a single Nvidia A100 GPU, despite benchmark claims that it rival…