This paper focuses on the piloting of the CGScholar AI Helper, a Generative AI (GenAI) assistant tool that aims to provide feedback on writing in high school contexts. The aim was to use GenAI to provide formative and summative feedback on students' texts in English Language Arts (ELA) and History. The trials discussed in this paper relate to Grade 11, a crucial learning phase when students are working towards college readiness. These trials took place in two very different schools in the Midwest of the United States, one in a low socio-economic background with low-performance outcomes and the other in a high socio-economic background with high-performance outcomes. The assistant tool used two main mechanisms "prompt engineering" based on participant teachers' assessment rubric and "fine-tuning" a Large Language Model (LLM) from a customized corpus of teaching materials using Retrieval Augmented Generation (RAG). This paper focuses on the CGScholar AI Helper's potential to enhance students' writing abilities and support teachers in ELA and other subject areas requiring written assignments.

Debate is a commonly used form of human communication catered towards problem-solving because of its efficiency. Debate fundamentally allows multiple viewpoints to be brought up in problem-solving, and for complex problems, each viewpoint opens a new path for problem-solving. In this work, we apply this concept to LLM decision-making by proposing town hall-style debate prompting (THDP), a prompting method that splices a language model into multiple personas that will debate one another to reach a conclusion. Our experimental pipeline varies both the number of personas and the personality types of each persona to find the optimum town hall size and personality for benchmark performance as measured by ZebraLogic bench, a reasoning-intensive benchmark characterized by both multiple-choice and fill-in-the-blank questions. Our experimental results demonstrate that a town hall size of 5 personas with LLM-determined personality types performs optimally on ZebraLogic, achieving a 13\% improvement over one-shot CoT baselines in per-cell accuracy in GPT-4o, 9% puzzle accuracy increase in Claude 3.5 Sonnet, and an improvement in hard puzzle accuracy from 10-15%.

In this study, the performance of various predictive models, including probabilistic baseline, CNN, LSTM, and finetuned LLMs, in forecasting merchant categories from financial transaction data have been evaluated. Utilizing datasets from Bank A for training and Bank B for testing, the superior predictive capabilities of the fine-tuned Mistral Instruct model, which was trained using customer data converted into natural language format have been demonstrated. The methodology of this study involves instruction fine-tuning Mistral via LoRA (LowRank Adaptation of Large Language Models) to adapt its vast pre-trained knowledge to the specific domain of financial transactions. The Mistral model significantly outperforms traditional sequential models, achieving higher F1 scores in the three key merchant categories of bank transaction data (grocery, clothing, and gas stations) that is crucial for targeted marketing campaigns. This performance is attributed to the model's enhanced semantic understanding and adaptability which enables it to better manage minority classes and predict transaction categories with greater accuracy. These findings highlight the potential of LLMs in predicting human behavior.

Reinforcement Learning from Human Feedback (RLHF) is commonly employed to tailor models to human preferences, especially to improve the safety of outputs from large language models (LLMs). Traditionally, this method depends on selecting preferred responses from pairs. However, due to the variability in human opinions and the challenges in directly comparing two responses, there is an increasing trend towards fine-grained annotation approaches that evaluate responses using multiple targeted metrics or rules. The challenge lies in efficiently choosing and applying these rules to handle the diverse range of preference data. In this paper, we propose a dynamic method that adaptively selects the most important rules for each response pair. We introduce a mathematical framework that utilizes the maximum discrepancy across paired responses and demonstrate theoretically that this approach maximizes the mutual information between the rule-based annotations and the underlying true preferences. We then train an 8B reward model using this adaptively labeled preference dataset and assess its efficacy using RewardBench. As of January 25, 2025, our model achieved the highest safety performance on the leaderboard, surpassing various larger models.

Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples input and output vocabularies to improve language modeling performance. Specifically, our approach scales up input vocabularies to leverage multi-gram tokens. Through extensive experiments, we uncover a log-linear relationship between input vocabulary size and training loss, demonstrating that larger input vocabularies consistently enhance model performance, regardless of model size. Using a large input vocabulary, we achieve performance comparable to double-sized baselines with no additional cost. Our findings highlight the importance of tokenization in scaling laws and provide practical insight for tokenizer design, paving the way for more efficient and powerful LLMs.

Background: The radiation oncology clinical practice involves many steps relying on the dynamic interplay of abundant text data. Large language models have displayed remarkable capabilities in processing complex text information. But their direct applications in specific fields like radiation oncology remain underexplored. Purpose: This study aims to investigate whether fine-tuning LLMs with domain knowledge can improve the performance on Task (1) treatment regimen generation, Task (2) treatment modality selection (photon, proton, electron, or brachytherapy), and Task (3) ICD-10 code prediction in radiation oncology. Methods: Data for 15,724 patient cases were extracted. Cases where patients had a single diagnostic record, and a clearly identifiable primary treatment plan were selected for preprocessing and manual annotation to have 7,903 cases of the patient diagnosis, treatment plan, treatment modality, and ICD-10 code. Each case was used to construct a pair consisting of patient diagnostics details and an answer (treatment regimen, treatment modality, or ICD-10 code respectively) for the supervised fine-tuning of these three tasks. Open source LLaMA2-7B and Mistral-7B models were utilized for the fine-tuning with the Low-Rank Approximations method. Accuracy and ROUGE-1 score were reported for the fine-tuned models and original models. Clinical evaluation was performed on Task (1) by radiation oncologists, while precision, recall, and F-1 score were evaluated for Task (2) and (3). One-sided Wilcoxon signed-rank tests were used to statistically analyze the results. Results: Fine-tuned LLMs outperformed original LLMs across all tasks with p-value <= 0.001. Clinical evaluation demonstrated that over 60% of the fine-tuned LLMs-generated treatment regimens were clinically acceptable. Precision, recall, and F1-score showed improved performance of fine-tuned LLMs.

Context-aware processing mechanisms have increasingly become a critical area of exploration for improving the semantic and contextual capabilities of language generation models. The Context-Aware Semantic Recomposition Mechanism (CASRM) was introduced as a novel framework designed to address limitations in coherence, contextual adaptability, and error propagation in large-scale text generation tasks. Through the integration of dynamically generated context vectors and attention modulation layers, CASRM enhances the alignment between token-level representations and broader contextual dependencies. Experimental evaluations demonstrated significant improvements in semantic coherence across multiple domains, including technical, conversational, and narrative text. The ability to adapt to unseen domains and ambiguous inputs was evaluated using a diverse set of test scenarios, highlighting the robustness of the proposed mechanism. A detailed computational analysis revealed that while CASRM introduces additional processing overhead, the gains in linguistic precision and contextual relevance outweigh the marginal increase in complexity. The framework also successfully mitigates error propagation in sequential tasks, improving performance in dialogue continuation and multi-step text synthesis. Additional investigations into token-level attention distribution emphasized the dynamic focus shifts enabled through context-aware enhancements. The findings suggest that CASRM offers a scalable and flexible solution for integrating contextual intelligence into existing language model architectures.

The United States' recent regulatory action against the Chinese-owned social video platform TikTok prompted mass migration to another Chinese app, the social platform "Rednote." Now, a generative artificial intelligence platform from the Chinese developer DeepSeek is exploding in popularity, posing a potential threat to US AI dominance and offering the latest evidence that moratoriums like the TikTok ban will not stop Americans from using Chinese-owned digital services. DeepSeek, an AI research lab created by a prominent Chinese hedge fund, recently gained popularity after releasing its latest open source generative AI model that easily competes with top US platforms like those developed by OpenAI. However, to help avoid US sanctions on hardware and software, DeepSeek created some clever workarounds when building its models. On Monday, DeepSeek's creators limited new sign-ups after claiming the app had been overrun with a "large-scale malicious attack."

The China-based AI company DeepSeek sent shock waves through both Silicon Valley and Wall Street by releasing an AI model that competes with the best US ones but was made at a fraction of the cost. Although the news has sparked a sell-off of tech stocks and prompted venture capitalist Marc Andreessen to describe the results as "AI's Sputnik moment" – the progress China has made on AI seems to have been a surprise to much of the tech world – this open-source AI is not as revolutionary…

The China-based AI company DeepSeek sent shock waves through both Silicon Valley and Wall Street by releasing an AI model that competes with the best US ones but was made at a fraction of the cost. Although the news has sparked a sell-off of tech stocks and prompted venture capitalist Marc Andreessen to describe the results as "AI's Sputnik moment", this open-source AI is not as revolutionary as it seems.