Search papers, labs, and topics across Lattice.
JevSpawn addresses the computational cost of token-by-token LLM agent inference by introducing a compositional policy that maps natural language task specifications to finite probabilistic exploration. This approach leverages parallel action spawning, feedback-driven branch selection, and representation revision to enable adaptive agentic behavior without requiring pre-defined action spaces. JevSpawn achieves improved task performance and faster navigation across eight benchmark tasks compared to seven agent baselines and a TypeSafe Jev variant, while reducing repeated generation through shared action structure and model prefixes.
LLM agents can now adapt their action spaces dynamically from natural language, bypassing the need for pre-defined finite fields and accelerating inference through compositional policies.
LLM agents generate intermediate reasoning and actions token by token, making extended interactions slow and computationally expensive. Jev-style models offer fast probabilistic predictions over finite fields, but require those fields to be specified in advance. This requirement limits autonomous task solving, where the available actions must be derived from natural language instructions and adapted through interaction. We introduce JevSpawn, a compositional policy that connects natural language task specifications to finite probabilistic exploration. Parallel action spawning is coupled with feedback driven branch selection, representation revision, and recovery from retained alternatives. Shared action structure and model prefixes reduce repeated generation and context computation without additional training. Evaluations on eight benchmark tasks against seven agent baselines and a TypeSafe Jev variant establish JevSpawn as a promising approach to structured agentic inference, with improved task performance and faster navigation.
