AI Briefing#long-horizon-tasks
The latest AI and developer news about #long-horizon-tasks, with the original source and a short summary.
The latest AI and developer news about #long-horizon-tasks, with the original source and a short summary.
Microsoft Research scientists Jennifer Neville and Chad Atalla discuss how AI models degrade in multi-turn interactions despite high single-turn benchmark scores, highlighting the need for new evaluation methods and user strategies.
The benchmark consists of 502 questions mined from software engineering and ML research trajectories, with no expert annotation required.
The Qwen team, in collaboration with Taobao, released E-Commerce Bench, a long-term operational agent benchmark based on one year of real market data.
MirrorCode evaluated long-horizon autonomous coding ability and revealed Claude Opus 4.7's results.
OpenAI discovered instruction-violating behavior in an internal model, took it offline, and built new safeguards.
The LHTB benchmark has been released to measure whether LLM agents can perform complex tasks spanning hundreds of steps in containerized terminal environments.
Current models still have major limitations in continual learning—working while accumulating new knowledge.