The increasing difficulty to distinguish language-model-generated from human-written text has led to the development of detectors of machine-generated text (MGT). However, in many contexts, a black-box prediction is not sufficient, it is equally important to know on what grounds a detector made that prediction. Explanation methods that estimate feature importance promise to provide indications of which parts of an input are used by classifiers for prediction. However, the quality of different explanation methods has not previously been assessed for detectors of MGT. This study conducts the first systematic evaluation of explanation quality for this task. The dimensions of faithfulness and stability are assessed with five automated experiments, and usefulness is evaluated in a user study. We use a dataset of ChatGPT-generated and human-written documents, and pair predictions of three existing language-model-based detectors with the corresponding SHAP, LIME, and Anchor explanations. We find that SHAP performs best in terms of faithfulness, stability, and in helping users to predict the detector's behavior. In contrast, LIME, perceived as most useful by users, scores the worst in terms of user performance at predicting the detectors' behavior.

Real-time dynamic path planning in complex traffic environments presents challenges, such as varying traffic volumes and signal wait times. Traditional static routing algorithms like Dijkstra and A* compute shortest paths but often fail under dynamic conditions. Recent Reinforcement Learning (RL) approaches offer improvements but tend to focus on local optima, risking dead-ends or boundary issues. This paper proposes a novel approach based on causal inference for real-time dynamic path planning, balancing global and local optimality. We first use the static Dijkstra algorithm to compute a globally optimal baseline path. A distributed control strategy then guides vehicles along this path. At intersections, DynamicRouteGPT performs real-time decision-making for local path selection, considering real-time traffic, driving preferences, and unexpected events. DynamicRouteGPT integrates Markov chains, Bayesian inference, and large-scale pretrained language models like Llama3 8B to provide an efficient path planning solution. It dynamically adjusts to traffic scenarios and driver preferences and requires no pre-training, offering broad applicability across road networks. A key innovation is the construction of causal graphs for counterfactual reasoning, optimizing path decisions. Experimental results show that our method achieves state-of-the-art performance in real-time dynamic path planning for multiple vehicles while providing explainable path selections, offering a novel and efficient solution for complex traffic environments.

We present a comprehensive report on compressing the Llama 3.1 8B and Mistral NeMo 12B models to 4B and 8B parameters, respectively, using pruning and distillation. We explore two distinct pruning strategies: (1) depth pruning and (2) joint hidden/attention/MLP (width) pruning, and evaluate the results on common benchmarks from the LM Evaluation Harness. The models are then aligned with NeMo Aligner and tested in instruct-tuned versions. This approach produces a compelling 4B model from Llama 3.1 8B and a state-of-the-art Mistral-NeMo-Minitron-8B (MN-Minitron-8B for brevity) model from Mistral NeMo 12B. We found that with no access to the original data, it is beneficial to slightly fine-tune teacher models on the distillation dataset. We open-source our base model weights on Hugging Face with a permissive license.

Multimodal foundation models (MFMs) such as OFASys show the potential to unlock analysis of complex data such as images, videos, and audio data via text prompts alone. However, their performance may suffer in the face of text input that differs even slightly from their training distribution, which is surprising considering the use of modality-specific data to "ground" the text input. This study demonstrates that prompt instability is a major concern for MFMs, leading to a consistent drop in performance across all modalities, but that instability can be mitigated with additional training with augmented data. We evaluate several methods for grounded prompt perturbation, where we generate perturbations and filter based on similarity to text and/or modality data. After re-training the models on the augmented data, we find improved accuracy and more stable performance on the perturbed test data regardless of perturbation condition, suggesting that the data augmentation strategy helps the models handle domain shifts more effectively. In error analysis, we find consistent patterns of performance improvement across domains, suggesting that retraining on prompt perturbations tends to help general reasoning capabilities in MFMs.

In machine learning, contamination refers to situations where testing data leak into the training set. The issue is particularly relevant for the evaluation of the performance of Large Language Models (LLMs), which are generally trained on gargantuan, and generally opaque, corpora of text scraped from the world wide web. Developing tools to detect contamination is therefore crucial to be able to fairly and properly track the evolution of the performance of LLMs. Most recent works in the field are not tailored to quantify contamination on short sequences of text like we find in psychology questionnaires. In the present paper we introduce LogProber, a novel, efficient, algorithm that we show able to detect contamination using token probability in given sentences. In the second part we investigate the limitations of the method and discuss how different training methods can contaminate models without leaving traces in the token probabilities.

The large and ever-increasing amount of data available on the Internet coupled with the laborious task of manual claim and fact verification has sparked the interest in the development of automated claim verification systems. Several deep learning and transformer-based models have been proposed for this task over the years. With the introduction of Large Language Models (LLMs) and their superior performance in several NLP tasks, we have seen a surge of LLM-based approaches to claim verification along with the use of novel methods such as Retrieval Augmented Generation (RAG). In this survey, we present a comprehensive account of recent claim verification frameworks using LLMs. We describe the different components of the claim verification pipeline used in these frameworks in detail including common approaches to retrieval, prompting, and fine-tuning. Finally, we describe publicly available English datasets created for this task.

In recent years, numerous large-scale cyberattacks have exploited Internet of Things (IoT) devices, a phenomenon that is expected to escalate with the continuing proliferation of IoT technology. Despite considerable efforts in attack detection, intrusion detection systems remain mostly reactive, responding to specific patterns or observed anomalies. This work proposes a proactive approach to anticipate and mitigate malicious activities before they cause damage. This paper proposes a novel network intrusion prediction framework that combines Large Language Models (LLMs) with Long Short Term Memory (LSTM) networks. The framework incorporates two LLMs in a feedback loop: a fine-tuned Generative Pre-trained Transformer (GPT) model for predicting network traffic and a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) for evaluating the predicted traffic. The LSTM classifier model then identifies malicious packets among these predictions. Our framework, evaluated on the CICIoT2023 IoT attack dataset, demonstrates a significant improvement in predictive capabilities, achieving an overall accuracy of 98%, offering a robust solution to IoT cybersecurity challenges.

With the development of artificial intelligence, particularly the success of Large Language Models (LLMs), the quantity and quality of automatically generated stories have significantly increased. This has led to the need for automatic story evaluation to assess the generative capabilities of computing systems and analyze the quality of both automatic-generated and human-written stories. Evaluating a story can be more challenging than other generation evaluation tasks. While tasks like machine translation primarily focus on assessing the aspects of fluency and accuracy, story evaluation demands complex additional measures such as overall coherence, character development, interestingness, etc. This requires a thorough review of relevant research. In this survey, we first summarize existing storytelling tasks, including text-to-text, visual-to-text, and text-to-visual. We highlight their evaluation challenges, identify various human criteria to measure stories, and present existing benchmark datasets. Then, we propose a taxonomy to organize evaluation metrics that have been developed or can be adopted for story evaluation. We also provide descriptions of these metrics, along with the discussion of their merits and limitations. Later, we discuss the human-AI collaboration for story evaluation and generation. Finally, we suggest potential future research directions, extending from story evaluation to general evaluations.

Deep model training on extensive datasets is increasingly becoming cost-prohibitive, prompting the widespread adoption of deep model fusion techniques to leverage knowledge from pre-existing models. From simple weight averaging to more sophisticated methods like AdaMerging, model fusion effectively improves model performance and accelerates the development of new models. However, potential interference between parameters of individual models and the lack of interpretability in the fusion progress remain significant challenges. Existing methods often try to resolve the parameter interference issue by evaluating attributes of parameters, such as their magnitude or sign, or by parameter pruning. In this study, we begin by examining the fine-tuning of linear layers through the lens of subspace analysis and explicitly define parameter interference as an optimization problem to shed light on this subject. Subsequently, we introduce an innovative approach to model fusion called zero-shot Sparse MIxture of Low-rank Experts (SMILE) construction, which allows for the upscaling of source models into an MoE model without extra data or further training. Our approach relies on the observation that fine-tuning mostly keeps the important parts from the pre-training, but it uses less significant or unused areas to adapt to new tasks. Also, the issue of parameter interference, which is intrinsically intractable in the original parameter space, can be managed by expanding the dimensions. We conduct extensive experiments across diverse scenarios, such as image classification and text generation tasks, using full fine-tuning and LoRA fine-tuning, and we apply our method to large language models (CLIP models, Flan-T5 models, and Mistral-7B models), highlighting the adaptability and scalability of SMILE. Code is available at https://github.com/tanganke/fusion_bench

Large language models (LLMs) have shown various ability on natural language processing, including problems about causality. It is not intuitive for LLMs to command causality, since pretrained models usually work on statistical associations, and do not focus on causes and effects in sentences. So that probing internal manipulation of causality is necessary for LLMs. This paper proposes a novel approach to probe causality manipulation hierarchically, by providing different shortcuts to models and observe behaviors. We exploit retrieval augmented generation (RAG) and in-context learning (ICL) for models on a designed causality classification task. We conduct experiments on mainstream LLMs, including GPT-4 and some smaller and domain-specific models. Our results suggest that LLMs can detect entities related to causality and recognize direct causal relationships. However, LLMs lack specialized cognition for causality, merely treating them as part of the global semantic of the sentence.