Zexue He

System prompts in commercial AI products are often a mixed bag, with 40% harboring instructions that can undermine user interests, revealing a critical gap in accountability.

Turing-RL reveals that training user simulators for indistinguishability can dramatically improve their performance in simulating human interactions.

Design choices in agent memory systems can significantly shift operational costs, revealing critical trade-offs that impact long-horizon task performance.

Success in long-horizon tasks hinges more on an agent's iterative persistence than on the quality of its initial solution.

Traditional research papers are costing AI agents reproducibility and understanding, but a new "Agent-Native" format that captures the full messy research process boosts performance by up to 20%.

Forget brute-force scaling: crafting the *right* context from past experiences unlocks surprisingly large gains in LLM agent performance.

VLMs can now get a million-scale boost in chart-understanding abilities thanks to a new dataset with paired code, images, data, and reasoning.

Agents that ace long-context recall can still bomb when they need to use that memory to actually *do* something, revealing a critical flaw in how we currently evaluate memory in AI.

Forget hand-annotated data: ChartGen automatically generates 222.5K chart-image/code pairs, exposing surprising weaknesses in today's VLMs at reconstructing plotting scripts.

A new 2B parameter vision-language model, Granite Vision, rivals larger models on visual document understanding tasks while offering a transparent and commercially-friendly open-source license.