LLM generated code often contains security issues. We address two key challenges in improving secure code generation. First, obtaining high quality training data covering a broad set of security issues is critical. To address this, we introduce a method for distilling a preference dataset of insecure and secure code pairs from frontier LLMs, along with a security reasoning that explains the issues and the fix. The key idea here is to make use of security knowledge sources to devise a systematic prompting strategy that ensures broad coverage. Second, aligning models to secure code requires focusing on localized regions of code. Direct preference optimization methods, like SimPO, are not designed to handle these localized differences and turn out to be ineffective. We address this with a new localized preference optimization algorithm that masks the security related tokens in both the winning (secure) and losing (insecure) responses. To prevent loss in code quality, we also add a regularizer. Evaluations show that both training on our dataset, DiSCo, and the new preference optimization algorithm, LPO, yield substantial reductions in code insecurity while also improving overall code quality. Code and dataset are available at https://github.com/StonyBrookNLP/disco-lpo.

Semantic Overlap Summarization (SOS) is a multi-document summarization task focused on extracting the common information shared cross alternative narratives which is a capability that is critical for trustworthy generation in domains such as news, law, and healthcare. We benchmark popular Large Language Models (LLMs) on SOS and introduce PrivacyPolicyPairs (3P), a new dataset of 135 high-quality samples from privacy policy documents, which complements existing resources and broadens domain coverage. Using the TELeR prompting taxonomy, we evaluate nearly one million LLM-generated summaries across two SOS datasets and conduct human evaluation on a curated subset. Our analysis reveals strong prompt sensitivity, identifies which automatic metrics align most closely with human judgments, and provides new baselines for future SOS research

One of the most important yet onerous tasks in the academic peer-reviewing process is composing meta-reviews, which involves assimilating diverse opinions from multiple expert peers, formulating one’s self-judgment as a senior expert, and then summarizing all these perspectives into a concise holistic overview to make an overall recommendation. This process is time-consuming and can be compromised by human factors like fatigue, inconsistency, missing tiny details, etc. Given the latest major developments in Large Language Models (LLMs), it is very compelling to rigorously study whether LLMs can help meta-reviewers perform this important task better. In this paper, we perform a case study with three popular LLMs, i.e., GPT-3.5, LLaMA2, and PaLM2, to assist meta-reviewers in better comprehending multiple experts’ perspectives by generating a controlled multi-perspective-summary (MPS) of their opinions. To achieve this, we prompt three LLMs with different types/levels of prompts based on the recently proposed TELeR taxonomy. Finally, we perform a detailed qualitative study of the MPSs generated by the LLMs and report our findings.

Sentence embeddings play a pivotal role in a wide range of NLP tasks, yet evaluating and interpreting these real-valued vectors remains an open challenge to date, especially in a task-free setting. To address this challenge, we introduce a novel task-free test bed for evaluating and interpreting sentence embeddings. Our test bed consists of five semantic similarity alignment criteria, namely, *semantic distinction, synonym replacement, antonym replacement, paraphrasing without negation, and sentence jumbling*. Using these criteria, we examined five classical (e.g., Sentence-BERT, Universal Sentence Encoder (USE), etc.) and eight LLM-induced sentence embedding techniques (e.g., LLaMA2, GPT-3, OLMo, etc.) to test whether their semantic similarity spaces align with what a human mind would naturally expect. Our extensive experiments with 13 different sentence encoders revealed that none of the studied embeddings aligned with all the five semantic similarity alignment criteria. Yet, most encoders performed highly on the SentEval dataset, a popular task-specific benchmark. This finding demonstrates a significant limitation of the current practice in sentence embedding evaluation and associated popular benchmarks, a critical issue that needs careful attention and reassessment by the NLP community. Finally, we conclude the paper by highlighting the utility of the proposed alignment-based test bed for analyzing sentence embeddings in a novel way, especially in a task-free setting.

This paper presents a benchmark dataset of Bangla word analogies for evaluating the quality of existing Bangla word embeddings. Despite being the 7th largest spoken language in the world, Bangla is still a low-resource language and popular NLP models often struggle to perform well on Bangla data sets. Therefore, developing a robust evaluation set is crucial for benchmarking and guiding future research on improving Bangla word embeddings, which is currently missing. To address this issue, we introduce a new evaluation set of 16,678 unique word analogies in Bangla as well as a translated and curated version of the original Mikolov dataset (10,594 samples) in Bangla. Our experiments with different state-of-the-art embedding models reveal that current Bangla word embeddings struggle to achieve high accuracy on both data sets, demonstrating a significant gap in multilingual NLP research.

In this paper, we conducted a comprehensive study with the latest Sentence Encoders and Large Language Models (LLMs) on the challenging task of “definition-wild zero-shot topic inference”, where users define or provide the topics of interest in real-time. Through extensive experimentation on seven diverse data sets, we observed that LLMs, such as ChatGPT-3.5 and PaLM, demonstrated superior generality compared to other LLMs, e.g., BLOOM and GPT-NeoX. Furthermore, Sentence-BERT, a BERT-based classical sentence encoder, outperformed PaLM and achieved performance comparable to ChatGPT-3.5.

While LLMs have shown great success in understanding and generating text in traditional conversational settings, their potential for performing ill-defined complex tasks is largely under-studied and yet to be benchmarked. However, conducting such benchmarking studies is challenging because of the large variations in LLMs’ performance when different prompt types/styles are used and different degrees of detail are provided in the prompts. To address this issue, this paper proposes a general taxonomy that can be used to design prompts with specific properties in order to perform a wide range of complex tasks. This taxonomy will allow future benchmarking studies to report the specific categories of prompts used as part of the study, enabling meaningful comparisons across different studies. Also, by establishing a common standard through this taxonomy, researchers will be able to draw more accurate conclusions about LLMs’ performance on a specific complex task.

As we begin to see low-powered computing paradigms (Neuromorphic Computing, Spiking Neural Networks, etc.) becoming more popular, learning binary word embeddings has become increasingly important for supporting NLP applications at the edge. Existing binary word embeddings are mostly derived from pretrained real-valued embeddings through different simple transformations, which often break the semantic consistency and the so-called “arithmetic” properties learned by the original, real-valued embeddings. This paper aims to address this limitation by introducing a new approach to learn binary embeddings from scratch, preserving the semantic relationships between words as well as the arithmetic properties of the embeddings themselves. To achieve this, we propose a novel genetic algorithm to learn the relationships between words from existing word analogy data-sets, carefully making sure that the arithmetic properties of the relationships are preserved. Evaluating our generated 16, 32, and 64-bit binary word embeddings on Mikolov’s word analogy task shows that more than 95% of the time, the best fit for the analogy is ranked in the top 5 most similar words in terms of cosine similarity.

Universal Sentence Encoder (USE) has gained much popularity recently as a general-purpose sentence encoding technique. As the name suggests, USE is designed to be fairly general and has indeed been shown to achieve superior performances for many downstream NLP tasks. In this paper, we present an interesting “negative” result on USE in the context of zero-shot text classification, a challenging task, which has recently gained much attraction. More specifically, we found some interesting cases of zero-shot text classification, where topic based inference outperformed USE-based inference in terms of F1 score. Further investigation revealed that USE struggles to perform well on data-sets with a large number of labels with high semantic overlaps, while topic-based classification works well for the same.

In this paper, we introduce an important yet relatively unexplored NLP task called Semantic Overlap Summarization (SOS), which entails generating a single summary from multiple alternative narratives which can convey the common information provided by those narratives. As no benchmark dataset is readily available for this task, we created one by collecting 2,925 alternative narrative pairs from the web and then, went through the tedious process of manually creating 411 different reference summaries by engaging human annotators. As a way to evaluate this novel task, we first conducted a systematic study by borrowing the popular ROUGE metric from text-summarization literature and discovered that ROUGE is not suitable for our task. Subsequently, we conducted further human annotations to create 200 document-level and 1,518 sentence-level ground-truth overlap labels. Our experiments show that the sentence-wise annotation technique with three overlap labels, i.e., Absent (A), Partially-Present (PP), and Present (P), yields a higher correlation with human judgment and higher inter-rater agreement compared to the ROUGE metric.

Recent work has introduced an important yet relatively under-explored NLP task called Semantic Overlap Summarization (SOS) that entails generating a summary from multiple alternative narratives which conveys the common information provided by those narratives. Previous work also published a benchmark dataset for this task by collecting 2,925 alternative narrative pairs from the web and manually annotating 411 different reference summaries by engaging human annotators. In this paper, we exclusively focus on the automated evaluation of the SOS task using the benchmark dataset. More specifically, we first use the popular ROUGE metric from text-summarization literature and conduct a systematic study to evaluate the SOS task. Our experiments discover that ROUGE is not suitable for this novel task and therefore, we propose a new sentence-level precision-recall style automated evaluation metric, called SEM-F1 (Semantic F1). It is inspired by the benefits of the sentence-wise annotation technique using overlap labels reported by the previous work. Our experiments show that the proposed SEM-F1 metric yields a higher correlation with human judgment and higher inter-rater agreement compared to the ROUGE metric.

Semantic Overlap Summarization (SOS) is a novel and relatively under-explored seq-to-seq task which entails summarizing common information from multiple alternate narratives. One of the major challenges for solving this task is the lack of existing datasets for supervised training. To address this challenge, we propose a novel data augmentation technique, which allows us to create large amount of synthetic data for training a seq-to-seq model that can perform the SOS task. Through extensive experiments using narratives from the news domain, we show that the models fine-tuned using the synthetic dataset provide significant performance improvements over the pre-trained vanilla summarization techniques and are close to the models fine-tuned on the golden training data; which essentially demonstrates the effectiveness of out proposed data augmentation technique for training seq-to-seq models on the SOS task.

It has been the norm for a long time to evaluate automated summarization tasks using the popular ROUGE metric. Although several studies in the past have highlighted the limitations of ROUGE, researchers have struggled to reach a consensus on a better alternative until today. One major limitation of the traditional ROUGE metric is the lack of semantic understanding (relies on direct overlap of n-grams). In this paper, we exclusively focus on the extractive summarization task and propose a semantic-aware nCG (normalized cumulative gain)-based evaluation metric (called Sem-nCG) for evaluating this task. One fundamental contribution of the paper is that it demonstrates how we can generate more reliable semantic-aware ground truths for evaluating extractive summarization tasks without any additional human intervention. To the best of our knowledge, this work is the first of its kind. We have conducted extensive experiments with this new metric using the widely used CNN/DailyMail dataset. Experimental results show that the new Sem-nCG metric is indeed semantic-aware, shows higher correlation with human judgement (more reliable) and yields a large number of disagreements with the original ROUGE metric (suggesting that ROUGE often leads to inaccurate conclusions also verified by humans).

Content of text data are often influenced by contextual factors which often evolve over time (e.g., content of social media are often influenced by topics covered in the major news streams). Existing language models do not consider the influence of such related evolving topics, and thus are not optimal. In this paper, we propose to incorporate such topical-influence into a language model to both improve its accuracy and enable cross-stream analysis of topical influences. Specifically, we propose a novel language model called Topical Influence Language Model (TILM), which is a novel extension of a neural language model to capture the influences on the contents in one text stream by the evolving topics in another related (or possibly same) text stream. Experimental results on six different text stream data comprised of conference paper titles show that the incorporation of evolving topical influence into a language model is beneficial and TILM outperforms multiple baselines in a challenging task of text forecasting. In addition to serving as a language model, TILM further enables interesting analysis of topical influence among multiple text streams.