Tag
Setting temperature to 0 does not guarantee deterministic tool calls in agents due to batched inference causing floating-point reduction order shifts, leading to token flips and different actions under load.
Introduces DMF, a deterministic memory framework for conversational AI agents that replaces LLM-based compression with classical NLP and mathematical scoring, achieving comparable accuracy to Mem0 while using zero tokens for memory preparation and up to 242× fewer tokens overall.
Summarizes a deterministic, constraint-based approach for building AI agents in regulated finance, where the LLM only generates prose, numbers are cryptographically sealed, and auditability is ensured through separated layers.
BotCircuits Agent is an open-source framework that introduces a Workflow-native Agent Loop architecture, splitting deterministic state-machine navigation from targeted LLM execution to reduce deviations and token costs.
ActiveGraph introduces a deterministic non-generative approach for evidence compilation before semantic memory, achieving 85.6% QA accuracy and 86.2% turn answer-in-context on LongMemEval-S.
Sponsio is an open-source deterministic contract layer that enforces tool-call boundaries and rules for LLM agents, addressing production reliability issues that prompt engineering and post-hoc auditing cannot fully solve.
The IMF published a formal note introducing the concept of agentic payments, framing the tension between probabilistic AI systems and deterministic payment infrastructure, and defining the shift from 'click to pay' to 'decide to pay'.
This paper introduces Ethical Hyper-Velocity (EHV), an architectural framework that combines conflict-free replicated data types and trusted execution environments to achieve sub-millisecond formal verification of AI governance policies in autonomous agentic systems, reducing governance latency from days to constant time.