Can AI Get Smarter And More Reliable? New Study Shows The Path Forward

New research reveals how AI can overcome its biggest challenges—like hallucinations and inefficiency—by integrating real-world knowledge, teaming up models, and evolving AI systems together to unlock smarter, more reliable solutions.

​​​​​​​Research: Knowledge-Empowered, Collaborative, and Co-Evolving AI Models: The Post-LLM Roadmap

A recent paper published in the journal Engineering delves into the future of artificial intelligence (AI) beyond large language models (LLMs). LLMs have made remarkable progress in multimodal tasks, yet they face limitations such as outdated information, hallucinations, inefficiency, and a lack of interpretability. Researchers explore three key directions to address these issues: knowledge empowerment, model collaboration, and model co-evolution.

Knowledge empowerment aims to integrate external knowledge into LLMs. This can be achieved through various methods, including incorporating knowledge into training objectives, instruction tuning, retrieval-augmented inference, and knowledge prompting. For example, some studies design knowledge-aware loss functions during pre-training, while others use retrieval-augmented generation to fetch relevant knowledge during inference dynamically. These techniques enhance LLMs' factual accuracy, reasoning capabilities, and interpretability.

Model collaboration focuses on leveraging the complementary strengths of different models. It includes strategies like model merging and collaboration based on various functional models. Model merging combines multiple models to improve performance, such as model ensembling and model fusion (e.g., the mixture of experts). LLMs can act as task managers in functional model collaboration, coordinating specialized small models. For instance, in image-generation tasks, LLMs can guide specialized models to better meet prompts' requirements.

Model co-evolution enables multiple models to evolve together. Under different types of heterogeneity-model, task, and data-various techniques have been proposed. Methods like parameter sharing, dual knowledge distillation, and hypernetwork-based parameter projection are used for model heterogeneity. In task heterogeneity, dual learning, adversarial learning, and model merging play important roles. When dealing with data heterogeneity, federated learning, and out-of-distribution knowledge distillation are key techniques. These methods enhance models' adaptability and ability to handle diverse tasks.

The post-LLM advancements have far-reaching impacts. In science, they help in hypothesis development by incorporating domain-specific knowledge. For example, AI models integrated with domain knowledge in meteorology can improve renewable energy forecasting. In engineering, they assist in problem formulation and solving. In society, they can be applied in areas like healthcare and traffic management.

The paper also points out several future research directions, including embodied AI, brain-like AI, non-transformer foundation models, and LLM-involved model generation. These areas hold great potential for further advancing AI capabilities. Integrating knowledge, collaboration, and co-evolution will be crucial in building more robust, efficient, and intelligent AI systems as AI continues to evolve.