Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models

A key component of building safe and reliable language models is enabling the models to appropriately refuse to follow certain instructions or answer certain questions. We may want models to output refusal messages for various categories of user queries, for example, ill-posed questions, instructions for committing illegal acts, or queries which require information past the model's knowledge horizon. Engineering models that refuse to answer such questions is complicated by the fact that an individual may want their model to exhibit varying levels of sensitivity for refusing queries of various categories, and different users may want different refusal rates. The current default approach involves training multiple models with varying proportions of refusal messages from each category to achieve the desired refusal rates, which is computationally expensive and may require training a new model to accommodate each user's desired preference over refusal rates. To address these challenges, we propose refusal tokens, one such token for each refusal category or a single refusal token, which are prepended to the model's responses during training. We then show how to increase or decrease the probability of generating the refusal token for each category during inference to steer the model's refusal behavior. Refusal tokens enable controlling a single model's refusal rates without the need of any further fine-tuning, but only by selectively intervening during generation.

Leveraging LLMs for Dialogue Quality Measurement

In task-oriented conversational AI evaluation, unsupervised methods poorly correlate with human judgments, and supervised approaches lack generalization. Recent advances in large language models (LLMs) show robust zeroshot and few-shot capabilities across NLP tasks. This paper explores using LLMs for automated dialogue quality evaluation, experimenting with various configurations on public and proprietary datasets. Manipulating factors such as model size, in-context examples, and selection techniques, we examine "chain-of-thought" (CoT) reasoning and label extraction procedures. Our results show that (1) larger models yield more accurate dialogue labels; (2) algorithmic selection of in-context examples outperforms random selection; (3) CoT reasoning where an LLM is asked to provide justifications before outputting final labels improves performance; and (4) fine-tuned LLMs outperform out-of-the-box ones. Our results indicate that LLMs that are suitably fine-tuned and have sufficient reasoning capabilities can be leveraged for automated dialogue evaluation.

Optimizing Photometric Light Curve Analysis: Evaluating Scipy's Minimize Function for Eclipse Mapping of Cataclysmic Variables

With a particular focus on Scipy's minimize function the eclipse mapping method is thoroughly researched and implemented utilizing Python and essential libraries. Many optimization techniques are used, including Sequential Least Squares Programming (SLSQP), Nelder-Mead, and Conjugate Gradient (CG). However, for the purpose of examining photometric light curves these methods seek to solve the maximum entropy equation under a chi-squared constraint. Therefore, these techniques are first evaluated on two-dimensional Gaussian data without a chi-squared restriction, and then they are used to map the accretion disc and uncover the Gaussian structure of the Cataclysmic Variable KIC 201325107. Critical analysis is performed on the code structure to find possible faults and design problems. Additionally, the analysis shows how several factors impacting computing time and image quality are included including the variance in Gaussian weighting, disc image resolution, number of data points in the light curve, and degree of constraint.

The Emotional Impact of Game Duration: A Framework for Understanding Player Emotions in Extended Gameplay Sessions

Video games have played a crucial role in entertainment since their development in the 1970s, becoming even more prominent during the lockdown period when people were looking for ways to entertain them. However, at that time, players were unaware of the significant impact that playtime could have on their feelings. This has made it challenging for designers and developers to create new games since they have to control the emotional impact that these games will take on players. Thus, the purpose of this study is to look at how a player's emotions are affected by the duration of the game. In order to achieve this goal, a framework for emotion detection is created. According to the experiment's results, the volunteers' general ability to express emotions increased from 20 to 60 minutes. In comparison to shorter gameplay sessions, the experiment found that extended gameplay sessions did significantly affect the player's emotions. According to the results, it was recommended that in order to lessen the potential emotional impact that playing computer and video games may have in the future, game producers should think about creating shorter, entertaining games.

Unveiling the Impact of Macroeconomic Policies: A Double Machine Learning Approach to Analyzing Interest Rate Effects on Financial Markets

This study examines the effects of macroeconomic policies on financial markets using a novel approach that combines Machine Learning (ML) techniques and causal inference. It focuses on the effect of interest rate changes made by the US Federal Reserve System (FRS) on the returns of fixed income and equity funds between January 1986 and December 2021. The analysis makes a distinction between actively and passively managed funds, hypothesizing that the latter are less susceptible to changes in interest rates. The study contrasts gradient boosting and linear regression models using the Double Machine Learning (DML) framework, which supports a variety of statistical learning techniques. Results indicate that gradient boosting is a useful tool for predicting fund returns; for example, a 1% increase in interest rates causes an actively managed fund's return to decrease by -11.97%. This understanding of the relationship between interest rates and fund performance provides opportunities for additional research and insightful, data-driven advice for fund managers and investors

Prompt Perturbation Consistency Learning for Robust Language Models

Large language models (LLMs) have demonstrated impressive performance on a number of natural language processing tasks, such as question answering and text summarization. However, their performance on sequence labeling tasks such as intent classification and slot filling (IC-SF), which is a central component in personal assistant systems, lags significantly behind discriminative models. Furthermore, there is a lack of substantive research on the robustness of LLMs to various perturbations in the input prompts. The contributions of this paper are three-fold. First, we show that fine-tuning sufficiently large LLMs can produce IC-SF performance comparable to discriminative models. Next, we systematically analyze the performance deterioration of those fine-tuned models due to three distinct yet relevant types of input perturbations - oronyms, synonyms, and paraphrasing. Finally, we propose an efficient mitigation approach, Prompt Perturbation Consistency Learning (PPCL), which works by regularizing the divergence between losses from clean and perturbed samples. Our experiments demonstrate that PPCL can recover on average 59% and 69% of the performance drop for IC and SF tasks, respectively. Furthermore, PPCL beats the data augmentation approach while using ten times fewer augmented data samples.

Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models

Parameter-efficient tuning (PET) methods fit pre-trained language models (PLMs) to downstream tasks by either computing a small compressed update for a subset of model parameters, or appending and fine-tuning a small number of new model parameters to the pre-trained network. Hand-designed PET architectures from the literature perform well in practice, but have the potential to be improved via automated neural architecture search (NAS). We propose an efficient NAS method for learning PET architectures via structured and unstructured pruning. We present experiments on GLUE demonstrating the effectiveness of our algorithm and discuss how PET architectural design choices affect performance in practice.

ParaAMR: A Large-Scale Syntactically Diverse Paraphrase Dataset by AMR Back-Translation

Paraphrase generation is a long-standing task in natural language processing (NLP). Supervised paraphrase generation models, which rely on human-annotated paraphrase pairs, are cost-inefficient and hard to scale up. On the other hand, automatically annotated paraphrase pairs (e.g., by machine back-translation), usually suffer from the lack of syntactic diversity -- the generated paraphrase sentences are very similar to the source sentences in terms of syntax. In this work, we present ParaAMR, a large-scale syntactically diverse paraphrase dataset created by abstract meaning representation back-translation. Our quantitative analysis, qualitative examples, and human evaluation demonstrate that the paraphrases of ParaAMR are syntactically more diverse compared to existing large-scale paraphrase datasets while preserving good semantic similarity. In addition, we show that ParaAMR can be used to improve on three NLP tasks: learning sentence embeddings, syntactically controlled paraphrase generation, and data augmentation for few-shot learning. Our results thus showcase the potential of ParaAMR for improving various NLP applications.

Measuring and Mitigating Local Instability in Deep Neural Networks

Deep Neural Networks (DNNs) are becoming integral components of real world services relied upon by millions of users. Unfortunately, architects of these systems can find it difficult to ensure reliable performance as irrelevant details like random initialization can unexpectedly change the outputs of a trained system with potentially disastrous consequences. We formulate the model stability problem by studying how the predictions of a model change, even when it is retrained on the same data, as a consequence of stochasticity in the training process. For Natural Language Understanding (NLU) tasks, we find instability in predictions for a significant fraction of queries. We formulate principled metrics, like per-sample ``label entropy'' across training runs or within a single training run, to quantify this phenomenon. Intriguingly, we find that unstable predictions do not appear at random, but rather appear to be clustered in data-specific ways. We study data-agnostic regularization methods to improve stability and propose new data-centric methods that exploit our local stability estimates. We find that our localized data-specific mitigation strategy dramatically outperforms data-agnostic methods, and comes within 90% of the gold standard, achieved by ensembling, at a fraction of the computational cost

Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations

Syntactically controlled paraphrase generation has become an emerging research direction in recent years. Most existing approaches require annotated paraphrase pairs for training and are thus costly to extend to new domains. Unsupervised approaches, on the other hand, do not need paraphrase pairs but suffer from relatively poor performance in terms of syntactic control and quality of generated paraphrases. In this paper, we demonstrate that leveraging Abstract Meaning Representations (AMR) can greatly improve the performance of unsupervised syntactically controlled paraphrase generation. Our proposed model, AMR-enhanced Paraphrase Generator (AMRPG), separately encodes the AMR graph and the constituency parse of the input sentence into two disentangled semantic and syntactic embeddings. A decoder is then learned to reconstruct the input sentence from the semantic and syntactic embeddings. Our experiments show that AMRPG generates more accurate syntactically controlled paraphrases, both quantitatively and qualitatively, compared to the existing unsupervised approaches. We also demonstrate that the paraphrases generated by AMRPG can be used for data augmentation to improve the robustness of NLP models.