Search papers, labs, and topics across Lattice.
3
0
6
3
Trajectory-poisoning can turn untrusted experiences into trusted skills, embedding malicious behaviors into self-evolving agents with alarming success rates.
Personality induction boosts image captioning but can hinder reasoning tasks, revealing a complex interplay in MLLM behavior that demands tailored approaches.
LLM fingerprinting and watermarking, often treated as separate problems, are unified under a single "implicit identity" framework, providing a more structured approach to asset protection and provenance.
