Tianyu Chen
PulseAugur coverage of Tianyu Chen — every cluster mentioning Tianyu Chen across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New LPS-Bench benchmark reveals safety flaws in AI agents
Researchers have developed LPS-Bench, a new benchmark designed to evaluate the safety awareness of computer-use agents (CUAs) in long-horizon planning tasks. The benchmark addresses the challenge of creating executable …
-
New framework enhances visual LLM agents with multi-step embedding retrieval
Researchers have developed VHOP, a new framework and benchmark designed to improve multi-step retrieval for visual LLM agents. This framework introduces VHOP-Router, a system that transforms standard embedding models in…
-
Apple unveils MT-EditFlow for improved multi-turn AI image editing
Apple researchers have developed MT-EditFlow, a new framework that uses reinforcement learning to improve multi-turn image editing. This approach addresses issues like error propagation and the all-or-nothing nature of …