Search papers, labs, and topics across Lattice.
This paper introduces G-Frame, an adaptive multi-agent framework that leverages Bayesian and team game principles to mitigate hallucinations in lightweight Large Language Models (LLMs) used in scientific domains. By enforcing structured reasoning and domain constraints, G-Frame synthesizes a specialized corpus of over 560,000 data points, enabling the training of the 7B model OmniChem. The model achieves performance parity with GPT-4o mini on custom benchmarks while demonstrating a significant 79.46% reduction in hallucinations, showcasing its potential in molecular design and synthesis planning.
A novel multi-agent framework reduces hallucinations in language models by 79.46%, enabling reliable reasoning in scientific applications.
The application of lightweight Large Language Models in rule-based scientific domains remains severely limited by their tendency to mimic linguistic patterns rather than reproduce axiomatic reasoning, causing frequent hallucinations. Here, we show that G-Frame, an adaptive multi-agent framework integrating Bayesian and team game principles, establishes an automated closed-loop for high-quality data synthesis and model training. By forcing the internalization of domain constraints through structured reasoning, we synthesized a specialized corpus of 363,045 chains-of-thought and 199,589 question-answer pairs. The resulting 7B model OmniChem achieves performance parity with GPT 4o mini on custom benchmarks and ChemBench while exhibiting a 79.46% reduction in hallucinations relative to its base architecture. We further demonstrate the advanced capabilities of OmniChem in molecular design and synthesis planning. This work establishes a scalable paradigm utilizing adaptive multi-agents to overcome inherent reasoning deficiencies, offering a feasible pathway for accelerating knowledge discovery in specialized scientific fields.
