Search papers, labs, and topics across Lattice.
6
0
7
6
AI governance policies can significantly enhance developer experience on GitHub, with transparency-focused strategies yielding better community and quality outcomes.
AutoTrace reveals that even state-of-the-art LLMs falter in causal reasoning, struggling to differentiate between vulnerable and safe code despite a robust new dataset.
Bug reports that work for humans can actually hinder AI agents, with localization cues being critical for repair success.
Agent personality profiles can drive performance differences of over 11 percentage points in multi-agent software engineering tasks, revealing a critical design factor often overlooked.
Developers find that mixing interaction types with GenAI can actually hinder productivity, challenging assumptions about tool integration in coding workflows.
SE researchers feel pressured to use GenAI, but are still wary of its trustworthiness, revealing a tension between perceived productivity gains and concerns about correctness and bias.
