Large Pre-Trained Language Models have demonstrated state-of-the-art performance in different downstream tasks, including dialogue state tracking and end-to-end response generation. Nevertheless, most of the publicly available datasets and benchmarks on task-oriented dialogues focus on written conversations. Consequently, the robustness of the developed models to spoken interactions is unknown. In this work, we have evaluated the performance of LLMs for spoken task-oriented dialogues on the DSTC11 test sets. Due to the lack of proper spoken dialogue datasets, we have automatically transcribed a development set of spoken dialogues with a state-of-the-art ASR engine. We have characterized the ASR-error types and their distributions and simulated these errors in a large dataset of dialogues. We report the intrinsic (perplexity) and extrinsic (human evaluation) performance of fine-tuned GPT-2 and T5 models in two subtasks of response generation and dialogue state tracking, respectively. The results show that LLMs are not robust to spoken noise by default, however, fine-tuning/training such models on a proper dataset of spoken TODs can result in a more robust performance.
相關內容
The unregulated use of LLMs can potentially lead to malicious consequences such as plagiarism, generating fake news, spamming, etc. Therefore, reliable detection of AI-generated text can be critical to ensure the responsible use of LLMs. Recent works attempt to tackle this problem either using certain model signatures present in the generated text outputs or by applying watermarking techniques that imprint specific patterns onto them. In this paper, we show that these detectors are not reliable in practical scenarios. In particular, we develop a recursive paraphrasing attack to apply on AI text, which can break a whole range of detectors, including the ones using the watermarking schemes as well as neural network-based detectors, zero-shot classifiers, and retrieval-based detectors. Our experiments include passages around 300 tokens in length, showing the sensitivity of the detectors even in the case of relatively long passages. We also observe that our recursive paraphrasing only degrades text quality slightly, measured via human studies, and metrics such as perplexity scores and accuracy on text benchmarks. Additionally, we show that even LLMs protected by watermarking schemes can be vulnerable against spoofing attacks aimed to mislead detectors to classify human-written text as AI-generated, potentially causing reputational damages to the developers. In particular, we show that an adversary can infer hidden AI text signatures of the LLM outputs without having white-box access to the detection method. Finally, we provide a theoretical connection between the AUROC of the best possible detector and the Total Variation distance between human and AI text distributions that can be used to study the fundamental hardness of the reliable detection problem for advanced language models. Our code is publicly available at //github.com/vinusankars/Reliability-of-AI-text-detectors.
Over the past few years, the abilities of large language models (LLMs) have received extensive attention, which have performed exceptionally well in complicated scenarios such as logical reasoning and symbolic inference. A significant factor contributing to this progress is the benefit of in-context learning and few-shot prompting. However, the reasons behind the success of such models using contextual reasoning have not been fully explored. Do LLMs have understand logical rules to draw inferences, or do they ``guess'' the answers by learning a type of probabilistic mapping through context? This paper investigates the reasoning capabilities of LLMs on two logical reasoning datasets by using counterfactual methods to replace context text and modify logical concepts. Based on our analysis, it is found that LLMs do not truly understand logical rules; rather, in-context learning has simply enhanced the likelihood of these models arriving at the correct answers. If one alters certain words in the context text or changes the concepts of logical terms, the outputs of LLMs can be significantly disrupted, leading to counter-intuitive responses. This work provides critical insights into the limitations of LLMs, underscoring the need for more robust mechanisms to ensure reliable logical reasoning in LLMs.
Policy gradient methods are widely adopted reinforcement learning algorithms for tasks with continuous action spaces. These methods succeeded in many application domains, however, because of their notorious sample inefficiency their use remains limited to problems where fast and accurate simulations are available. A common way to improve sample efficiency is to modify their objective function to be computable from off-policy samples without importance sampling. A well-established off-policy objective is the excursion objective. This work studies the difference between the excursion objective and the traditional on-policy objective, which we refer to as the on-off gap. We provide the first theoretical analysis showing conditions to reduce the on-off gap while establishing empirical evidence of shortfalls arising when these conditions are not met.
Large Language Models (LLMs) have shown impressive proficiency in code generation. Nonetheless, similar to human developers, these models might generate code that contains security vulnerabilities and flaws. Writing secure code remains a substantial challenge, as vulnerabilities often arise during interactions between programs and external systems or services, such as databases and operating systems. In this paper, we propose a novel approach, Feedback-Driven Solution Synthesis (FDSS), designed to explore the use of LLMs in receiving feedback from Bandit, which is a static code analysis tool, and then the LLMs generate potential solutions to resolve security vulnerabilities. Each solution, along with the vulnerable code, is then sent back to the LLM for code refinement. Our approach shows a significant improvement over the baseline and outperforms existing approaches. Furthermore, we introduce a new dataset, PythonSecurityEval, collected from real-world scenarios on Stack Overflow to evaluate the LLMs' ability to generate secure code. Code and data are available at \url{//github.com/Kamel773/LLM-code-refine}
Network pruning can reduce the computation cost of deep neural network (DNN) models. However, sparse models often produce randomly-distributed weights to maintain accuracy, leading to irregular computations. Consequently, unstructured sparse models cannot achieve meaningful speedup on commodity hardware built for dense matrix computations. Accelerators are usually modified or designed with structured sparsity-optimized architectures for exploiting sparsity. For example, the Ampere architecture introduces a sparse tensor core, which adopts the 2:4 sparsity pattern. We propose a pruning method that builds upon the insight that matrix multiplication generally breaks the large matrix into multiple smaller tiles for parallel execution. We present the tile-wise sparsity pattern, which maintains a structured sparsity pattern at the tile level for efficient execution but allows for irregular pruning at the global scale to maintain high accuracy. In addition, the tile-wise sparsity is implemented at the global memory level, and the 2:4 sparsity executes at the register level inside the sparse tensor core. We can combine these two patterns into a tile-vector-wise (TVW) sparsity pattern to explore more fine-grained sparsity and further accelerate the sparse DNN models. We evaluate the TVW on the GPU, achieving averages of $1.85\times$, $2.75\times$, and $22.18\times$ speedups over the dense model, block sparsity, and unstructured sparsity.
Work on instruction-tuned Large Language Models (LLMs) has used automatic methods based on text overlap and LLM judgments as cost-effective alternatives to human evaluation. In this paper, we study the reliability of such methods across a broad range of tasks and in a cross-lingual setting. In contrast to previous findings, we observe considerable variability in correlations between automatic methods and human evaluators when scores are differentiated by task type. Specifically, the widely-used ROUGE-L metric strongly correlates with human judgments for short-answer English tasks but is unreliable in free-form generation tasks and cross-lingual transfer. The effectiveness of GPT-4 as an evaluator depends on including reference answers when prompting for assessments, which can lead to overly strict evaluations in free-form generation tasks. In summary, we find that, while automatic evaluation methods can approximate human judgements under specific conditions, their reliability is highly context-dependent. Our findings enhance the understanding of how automatic methods should be applied and interpreted when developing and evaluating instruction-tuned LLMs.
Chain-of-Thought (CoT) prompting has marked a significant advancement in enhancing the reasoning capabilities of large language models (LLMs). Previous studies have developed various extensions of CoT, which focus primarily on enhancing end-task performance. In addition, there has been research on assessing the quality of reasoning chains in CoT. This raises an intriguing question: Is it possible to predict the accuracy of LLM outputs by scrutinizing the reasoning chains they generate? To answer this research question, we introduce a benchmark, R2PE, designed specifically to explore the relationship between reasoning chains and performance in various reasoning tasks spanning five different domains. This benchmark aims to measure the falsehood of the final output of LLMs based on the reasoning steps. To make full use of information in multiple reasoning chains, we propose the process discernibility score (PDS) framework that beats the answer-checking baseline by a large margin. Concretely, this resulted in an average of 5.1% increase in the F1 score across all 45 subsets within R2PE. We further demonstrate our PDS's efficacy in advancing open-domain QA accuracy. Data and code are available at //github.com/XinXU-USTC/R2PE.
In the field of robotics and automation, navigation systems based on Large Language Models (LLMs) have recently shown impressive performance. However, the security aspects of these systems have received relatively less attention. This paper pioneers the exploration of vulnerabilities in LLM-based navigation models in urban outdoor environments, a critical area given the technology's widespread application in autonomous driving, logistics, and emergency services. Specifically, we introduce a novel Navigational Prompt Suffix (NPS) Attack that manipulates LLM-based navigation models by appending gradient-derived suffixes to the original navigational prompt, leading to incorrect actions. We conducted comprehensive experiments on an LLMs-based navigation model that employs various LLMs for reasoning. Our results, derived from the Touchdown and Map2Seq street-view datasets under both few-shot learning and fine-tuning configurations, demonstrate notable performance declines across three metrics in the face of both white-box and black-box attacks. These results highlight the generalizability and transferability of the NPS Attack, emphasizing the need for enhanced security in LLM-based navigation systems. As an initial countermeasure, we propose the Navigational Prompt Engineering (NPE) Defense strategy, concentrating on navigation-relevant keywords to reduce the impact of adversarial suffixes. While initial findings indicate that this strategy enhances navigational safety, there remains a critical need for the wider research community to develop stronger defense methods to effectively tackle the real-world challenges faced by these systems.
Adversarial attack is a technique for deceiving Machine Learning (ML) models, which provides a way to evaluate the adversarial robustness. In practice, attack algorithms are artificially selected and tuned by human experts to break a ML system. However, manual selection of attackers tends to be sub-optimal, leading to a mistakenly assessment of model security. In this paper, a new procedure called Composite Adversarial Attack (CAA) is proposed for automatically searching the best combination of attack algorithms and their hyper-parameters from a candidate pool of \textbf{32 base attackers}. We design a search space where attack policy is represented as an attacking sequence, i.e., the output of the previous attacker is used as the initialization input for successors. Multi-objective NSGA-II genetic algorithm is adopted for finding the strongest attack policy with minimum complexity. The experimental result shows CAA beats 10 top attackers on 11 diverse defenses with less elapsed time (\textbf{6 $\times$ faster than AutoAttack}), and achieves the new state-of-the-art on $l_{\infty}$, $l_{2}$ and unrestricted adversarial attacks.
Graph Neural Networks (GNNs) for representation learning of graphs broadly follow a neighborhood aggregation framework, where the representation vector of a node is computed by recursively aggregating and transforming feature vectors of its neighboring nodes. Many GNN variants have been proposed and have achieved state-of-the-art results on both node and graph classification tasks. However, despite GNNs revolutionizing graph representation learning, there is limited understanding of their representational properties and limitations. Here, we present a theoretical framework for analyzing the expressive power of GNNs in capturing different graph structures. Our results characterize the discriminative power of popular GNN variants, such as Graph Convolutional Networks and GraphSAGE, and show that they cannot learn to distinguish certain simple graph structures. We then develop a simple architecture that is provably the most expressive among the class of GNNs and is as powerful as the Weisfeiler-Lehman graph isomorphism test. We empirically validate our theoretical findings on a number of graph classification benchmarks, and demonstrate that our model achieves state-of-the-art performance.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191