[Paper Review] BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence

Jason Lee, Xuezhi Wang|arXiv (Cornell University)|Jan 28, 2022

Topic Modeling7 references4,210 citations

TL;DR

This paper investigates chain-of-thought prompting as a simple method to elicit multi-step reasoning in large language models, showing emergent reasoning abilities with scale across arithmetic, commonsense, and symbolic tasks. It demonstrates large gains on benchmarks like GSM8K without fine-tuning.

ABSTRACT

Below is a detailed introduction in English for the document: Detailed Introduction The document “BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence (Results + Code)” represents an innovative research work led by Dr. Francesco Bulla, in collaboration with Stephanie Ewelu. At its core, the document introduces a revolutionary paradigm that overcomes the traditional limitations imposed by tokenization. Instead of breaking down input into discrete tokens, the proposed approach is based on a continuous processing of data, allowing the AI model to maintain a “digital DNA” that remains self-consistent throughout its operations. The document is structured into several sections that cover both the theoretical foundations and the technical implementation details. It provides the complete BNAI code that integrates the NO-TOKEN module, which is capable of operating on different hardware architectures. In particular, the study presents experiments conducted on high-performance GPU setups (such as the NVIDIA RTX 3090) as well as CPU-only environments (using a 16-core processor). The experimental results, expressed in terms of accuracy, latency, and loss metrics, clearly demonstrate that the token-free model not only offers performance levels comparable to the traditional token-based systems (for example, BERT-base) but also presents significant advantages in efficiency and scalability, especially in resource-constrained settings. A further innovative aspect of the work is the concept of “digital DNA”. This notion translates into an identity encoding for the model, where ethics, adaptability, and security are integrated directly into the learning process. This enables the system not only to continuously evolve and learn from its mistakes but also to preserve a form of coherent digital identity—a key element for the development of autonomous, traceable, and accountable AI systems in the future. Moreover, the document not only presents detailed experimental results but also offers a comparative analysis with previous methodologies. It highlights, for instance, a reduction in latency of approximately 20% when compared to traditional tokenized methods, alongside achieving similar levels of accuracy. This approach, therefore, paves the way for continuous learning and evolution in AI—emphasizing the fusion of performance with intrinsic ethical and security features, all of which set the stage for the future of autonomous AI development. Finally, the work outlines a roadmap for further developments, suggesting additional benchmarks and comparisons with state-of-the-art methods like ByT5, as well as approaches based on wavelets. The document invites the scientific community to engage in constructive discussion, fostering a collaborative environment toward an AI evolution that is not only technically advanced but also ethically and responsibly grounded. Conclusions and References Main ReferencesZenodo DOI for this document:10.5281/zenodo.15185971 Related Works Articles available on arXiv, Nature, ScienceDirect, IEEE, etc., covering topics such as:Continuous learning, bias and fairness, hypernets, autopoiesis in artificial intelligence, tokenization problems, wavelet-based embeddings, and more.A complete bibliography is available in the PDF.How to cite (example)Bulla, F., & Ewelu, S. (2025). BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence. Zenodo.https://doi.org/10.5281/zenodo.14894878 or cite https://doi.org/10.5281/zenodo.15185971 Final SummaryThis document presents a transformative vision for AI by replacing token-based constraints with an AI model that preserves a self-consistent “digital DNA” (through BNAI). It details robust mathematical computations and real-world tests, demonstrating how AI can self-govern (MIND-UNITY) and continuously process data (NO-TOKEN). By integrating ethics and security as intrinsic factors, it paves the way for the evolution of autonomous, traceable, and responsible AI, offering a roadmap for collaborative and open development in which models retain their identity and adapt securely.

Motivation & Objective

• Motivate how reasoning can emerge in large language models via natural language intermediates (chain-of-thought).
• Demonstrate that chain-of-thought prompting yields substantial performance gains on arithmetic, commonsense, and symbolic tasks.
• Show robustness of the approach across annotators, exemplars, and model families.
• Highlight the emergent nature of chain-of-thought ability as model scale increases.

Proposed method

• Prompt models with few-shot triples: (input, chain of thought, output) to elicit intermediate reasoning steps.
• Evaluate across multiple large language models (GPT-3, LaMDA, PaLM, UL2, Codex) and scales (from hundreds of millions to hundreds of billions of parameters).
• Compare chain-of-thought prompting to standard prompting and finetuned baselines on diverse benchmarks.
• Analyze exemplar robustness, including annotator style, exemplar order, and varying exemplar sets.
• Conduct ablations to separate the effect of reasoning steps from equation generation or mere increased computation.
• Assess both in-domain and out-of-domain (OOD) generalization for symbolic tasks.

Experimental results

Research questions

• RQ1 Does chain-of-thought prompting unlock multi-step reasoning in large language models?
• RQ2 How does model scale affect the effectiveness of chain-of-thought prompting?
• RQ3 Can chain-of-thought prompting improve performance on arithmetic, commonsense, and symbolic reasoning tasks without fine-tuning?
• RQ4 Is the approach robust to different exemplars, annotators, and model choices?
• RQ5 Does chain-of-thought prompting generalize to out-of-domain or longer reasoning sequences?

Key findings

• Chain-of-thought prompting is an emergent capability that appears with model sizes around ~100B parameters.
• It yields large performance gains on arithmetic tasks (e.g., GSM8K) compared to standard prompting, approaching or surpassing finetuned baselines.
• The gains extend to commonsense reasoning and symbolic tasks, with notable improvements for the PaLM 540B model.
• Chain-of-thought prompts show robustness across annotators, exemplar sets, and model families, though results vary with exemplars.
• Scaling to PaLM 540B and LaMDA/GPT-3 variants demonstrates competitive or state-of-the-art performance on several benchmarks (e.g., GSM8K, SVAMP, MAWPS, StrategyQA).
• Ablations indicate that the benefits arise from the reasoning process itself rather than merely increased computation or post-hoc use of the chain of thought.
• Out-of-domain and longer sequence reasoning can be better solved with chain-of-thought prompting at sufficient scale, suggesting improved length generalization.