Bro Saw Two Photos and Made Up Lore
A developer built UNSEEN, a walking game that uses a local open-weight vision model (Qwen3-VL-2B-Instruct via Ollama) only to describe photos as structured data, while deterministic code — an "evidenc…
A developer built UNSEEN, a walking game that uses a local open-weight vision model (Qwen3-VL-2B-Instruct via Ollama) only to describe photos as structured data, while deterministic code — an "evidenc…
A developer built MemoryBox, an open-weight AI pipeline that lets users search a personal photo archive using natural-language descriptions of moments rather than filenames or dates. The system uses Q…
Researchers introduced MindTopo, a benchmark of 11,030 instances across 13 procedurally generated task types that tests topological intuition in foundation models across five properties: continuity, s…
NVIDIA has demonstrated an autoresearch workflow using RL agent skills and the NeMo framework, where a frontier coding agent (Codex with GPT 5.5) autonomously set up a NeMo RL and NeMo Gym stack, crea…