Recent observations have underscored a disparity between the inflated benchmark scores and the actual performance of LLMs, raising concerns about potential contamination of evaluation benchmarks. This issue is especially critical for closed-source models and certain open-source models where training data transparency is lacking. In this paper we study data contamination by proposing two methods tailored for both open-source and proprietary LLMs. We first introduce a retrieval-based system to explore potential overlaps between evaluation benchmarks and pretraining corpora. We further present a novel investigation protocol named Testset Slot Guessing (TS-Guessing), applicable to both open and proprietary models. This approach entails masking a wrong answer in a multiple-choice question and prompting the model to fill in the gap. Additionally, it involves obscuring an unlikely word in an evaluation example and asking the model to produce it. We find that certain commercial LLMs could surprisingly guess the missing option in various test sets. Specifically, in the MMLU benchmark, ChatGPT and GPT-4 demonstrated an exact match rate of 52% and 57%, respectively, in guessing the missing options in benchmark test data. We hope these results underscore the need for more robust evaluation methodologies and benchmarks in the field.

Despite the remarkable capabilities of Large Language Models (LLMs) like GPT-4, producing complex, structured tabular data remains challenging. Our study assesses LLMs’ proficiency in structuring tables and introduces a novel fine-tuning method, cognizant of data structures, to bolster their performance. We unveil Struc-Bench, a comprehensive benchmark featuring prominent LLMs (GPT-NeoX-20B, GPT-3.5, GPT-4, and Vicuna), which spans text tables, HTML, and LaTeX formats. Our proposed FormatCoT aids in crafting format-specific instructions from the intended outputs to populate this benchmark. Addressing the gap in task-centered evaluation, we propose two innovative metrics, P-Score (Prompting Score) and H-Score (Heuristical Score), to more accurately gauge LLM performance. Our experiments show that applying our structure-aware fine-tuning to LLaMA-7B leads to substantial performance gains, outshining its LLM counterparts across most measures. In-depth error analysis and creating an ability map across six dimensions, coverage, formatting, reasoning, comprehension, pragmatics, and hallucination, highlight areas for future enhancements and suggest forthcoming research trajectories. Our code and models can be found at https://github.com/gersteinlab/Struc-Bench.

In the rapidly evolving landscape of medical research, accurate and concise summarization of clinical studies is crucial to support evidence-based practice. This paper presents a novel approach to clinical studies summarization, leveraging reinforcement learning to enhance factual consistency and align with human annotator preferences. Our work focuses on two tasks: Conclusion Generation and Review Generation. We train a CONFIT summarization model that outperforms GPT-3 and previous state-of-the-art models on the same datasets and collects expert and crowd-worker annotations to evaluate the quality and factual consistency of the generated summaries. These annotations enable us to measure the correlation of various automatic metrics, including modern factual evaluation metrics like QAFactEval, with human-assessed factual consistency. By employing top-correlated metrics as objectives for a reinforcement learning model, we demonstrate improved factuality in generated summaries that are preferred by human annotators.

This paper presents our contribution to the MEDIQA-2023 Dialogue2Note shared task, encompassing both subtask A and subtask B. We approach the task as a dialogue summarization problem and implement two distinct pipelines: (a) a fine-tuning of a pre-trained dialogue summarization model and GPT-3, and (b) few-shot in-context learning (ICL) using a large language model, GPT-4. Both methods achieve excellent results in terms of ROUGE-1 F1, BERTScore F1 (deberta-xlarge-mnli), and BLEURT, with scores of 0.4011, 0.7058, and 0.5421, respectively. Additionally, we predict the associated section headers using RoBERTa and SciBERT based classification models. Our team ranked fourth among all teams, while each team is allowed to submit three runs as part of their submission. We also utilize expert annotations to demonstrate that the notes generated through the ICL GPT-4 are better than all other baselines. The code for our submission is available.