A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Erik Nijkamp, Bo Pang, Hiroaki Hayashi, et al.
CodeGen introduces a 16.1B parameter LLM for program synthesis that improves code generation through multi-turn prompting.
Baohao Liao, Hanze Dong, Christof Monz, et al.
ReOPD enables efficient multi-turn on-policy distillation for LLM agents by reusing teacher trajectories as replayed prefixes, avoiding costly environment interactions.
Jihoon Tack, Philippe Laban, Jennifer Neville
Introduces a framework to convert static tasks into dynamic multi-turn conversations, revealing that LLMs fail to track evolving user intent despite strong static performance.
Unknown
Agentboard introduces an analytical evaluation board for multi-turn LLM agents, focusing on their capability for analytic evaluation to advance agent development.
Xu Li, Simon Yu, Minzhou Pan, et al.
Introduces MTAgentRisk, the first multi-turn safety benchmark for tool-using agents, revealing a 16% average increase in attack success rate across multiple turns.