Large language models (LLMs) have excelled in various natural language processing tasks, but challenges in interpretability and trustworthiness persist, limiting their use in high-stakes fields. Causal discovery offers a promising approach to improve transparency and reliability. However, current evaluations are often one-sided and lack assessments focused on interpretability performance. Additionally, these evaluations rely on synthetic data and lack comprehensive assessments of real-world datasets. These lead to promising methods potentially being overlooked. To address these issues, we propose a flexible evaluation framework with metrics for evaluating differences in causal structures and causal effects, which are crucial attributes that help improve the interpretability of LLMs. We introduce the Open Causal Discovery Benchmark (OCDB), based on real data, to promote fair comparisons and drive optimization of algorithms. Additionally, our new metrics account for undirected edges, enabling fair comparisons between Directed Acyclic Graphs (DAGs) and Completed Partially Directed Acyclic Graphs (CPDAGs). Experimental results show significant shortcomings in existing algorithms' generalization capabilities on real data, highlighting the potential for performance improvement and the importance of our framework in advancing causal discovery techniques.

In recent years, machine learning (ML) techniques have created numerous opportunities for intelligent mobile networks and have accelerated the automation of network operations. However, complex network tasks may involve variables and considerations even beyond the capacity of traditional ML algorithms. On the other hand, large language models (LLMs) have recently emerged, demonstrating near-human-level performance in cognitive tasks across various fields. However, they remain prone to hallucinations and often lack common sense in basic tasks. Therefore, they are regarded as assistive tools for humans. In this work, we propose the concept of "generative AI-in-the-loop" and utilize the semantic understanding, context awareness, and reasoning abilities of LLMs to assist humans in handling complex or unforeseen situations in mobile communication networks. We believe that combining LLMs and ML models allows both to leverage their respective capabilities and achieve better results than either model alone. To support this idea, we begin by analyzing the capabilities of LLMs and compare them with traditional ML algorithms. We then explore potential LLM-based applications in line with the requirements of next-generation networks. We further examine the integration of ML and LLMs, discussing how they can be used together in mobile networks. Unlike existing studies, our research emphasizes the fusion of LLMs with traditional ML-driven next-generation networks and serves as a comprehensive refinement of existing surveys. Finally, we provide a case study to enhance ML-based network intrusion detection with synthesized data generated by LLMs. Our case study further demonstrates the advantages of our proposed idea.

Predictable behavior from scaling advanced AI systems is an extremely desirable property. Although a well-established literature exists on how pretraining performance scales, the literature on how particular downstream capabilities scale is significantly muddier. In this work, we take a step back and ask: why has predicting specific downstream capabilities with scale remained elusive? While many factors are certainly responsible, we identify a new factor that makes modeling scaling behavior on widely used multiple-choice question-answering benchmarks challenging. Using five model families and twelve well-established multiple-choice benchmarks, we show that downstream performance is computed from negative log likelihoods via a sequence of transformations that progressively degrade the statistical relationship between performance and scale. We then reveal the mechanism causing this degradation: downstream metrics require comparing the correct choice against a small number of specific incorrect choices, meaning accurately predicting downstream capabilities requires predicting not just how probability mass concentrates on the correct choice with scale, but also how probability mass fluctuates on specific incorrect choices with scale. We empirically study how probability mass on the correct choice co-varies with probability mass on incorrect choices with increasing compute, suggesting that scaling laws for incorrect choices might be achievable. Our work also explains why pretraining scaling laws are commonly regarded as more predictable than downstream capabilities and contributes towards establishing scaling-predictable evaluations of frontier AI models.

When using language models (LMs) to solve complex problems, humans might struggle to understand the LM-generated solutions and repair the flawed ones. To assist humans in repairing them, we propose to automatically decompose complex solutions into multiple simpler pieces that correspond to specific subtasks. We introduce a novel objective for learning task decomposition, termed assistive value (AssistV), which measures the feasibility and speed for humans to repair the decomposed solution. We collect a dataset of human repair experiences on different decomposed solutions. Utilizing the collected data as in-context examples, we then learn to critique, refine, and rank decomposed solutions to improve AssistV. We validate our method under competitive programming problems: under 177 hours of human study, our method enables non-experts to solve 33.3\% more problems, speeds them up by 3.3x, and empowers them to match unassisted experts.

A good teacher should not only be knowledgeable; but should be able to communicate in a way that the student understands -- to share the student's representation of the world. In this work, we integrate insights from machine teaching and pragmatic communication with the burgeoning literature on representational alignment to characterize a utility curve defining a relationship between representational alignment and teacher capability for promoting student learning. To explore the characteristics of this utility curve, we design a supervised learning environment that disentangles representational alignment from teacher accuracy. We conduct extensive computational experiments with machines teaching machines, complemented by a series of experiments in which machines teach humans. Drawing on our findings that improved representational alignment with a student improves student learning outcomes (i.e., task accuracy), we design a classroom matching procedure that assigns students to teachers based on the utility curve. If we are to design effective machine teachers, it is not enough to build teachers that are accurate -- we want teachers that can align, representationally, to their students too.

Controlling the attribute intensity of text generation is crucial across scenarios (e.g., writing conciseness, chatting emotion, and explanation clarity). The remarkable capabilities of large language models (LLMs) have revolutionized text generation, prompting us to explore such \emph{smooth control} of LLM generation. Specifically, we propose metrics to assess the range, calibration, and consistency of the generated text's attribute intensity in response to varying control values, as well as its relevance to the intended context. To quantify the attribute intensity and context relevance, we propose an effective evaluation framework leveraging the Elo rating system and GPT4, both renowned for their robust alignment with human judgment. We look into two viable training-free methods for achieving smooth control of LLMs: (1) Prompting with semantic shifters, and (2) Modifying internal model representations. The evaluations of these two methods are conducted on $5$ different attributes with various models. Our code and dataset can be obtained from \url{https://github.com/ShangDataLab/Smooth-Control}.

This study addresses a crucial challenge in instance-based relation extraction using text generation models: end-to-end training in target relation extraction task is not applicable to retrievers due to the non-differentiable nature of instance selection. We propose a novel End-to-end TRAinable Soft K-nearest neighbor retriever (ETRASK) by the neural prompting method that utilizes a soft, differentiable selection of the $k$ nearest instances. This approach enables the end-to-end training of retrievers in target tasks. On the TACRED benchmark dataset with a low-resource setting where the training data was reduced to 10\%, our method achieved a state-of-the-art F1 score of 71.5\%. Moreover, ETRASK consistently improved the baseline model by adding instances for all settings. These results highlight the efficacy of our approach in enhancing relation extraction performance, especially in resource-constrained environments. Our findings offer a promising direction for future research with extraction and the broader application of text generation in natural language processing.

Reasoning about actions and change (RAC) has historically driven the development of many early AI challenges, such as the frame problem, and many AI disciplines, including non-monotonic and commonsense reasoning. The role of RAC remains important even now, particularly for tasks involving dynamic environments, interactive scenarios, and commonsense reasoning. Despite the progress of Large Language Models (LLMs) in various AI domains, their performance on RAC is underexplored. To address this gap, we introduce a new benchmark, ActionReasoningBench, encompassing 13 domains and rigorously evaluating LLMs across eight different areas of RAC. These include - Object Tracking, Fluent Tracking, State Tracking, Action Executability, Effects of Actions, Numerical RAC, Hallucination Detection, and Composite Questions. Furthermore, we also investigate the indirect effect of actions due to ramification constraints for every domain. Finally, we evaluate our benchmark using open-sourced and commercial state-of-the-art LLMs, including GPT-4o, Gemini-1.0-Pro, Llama2-7b-chat, Llama2-13b-chat, Llama3-8b-instruct, Gemma-2b-instruct, and Gemma-7b-instruct. Our findings indicate that these models face significant challenges across all categories included in our benchmark.

Multi-modal contrastive models such as CLIP (Radford et al. (2021)) achieve stateof-the-art performance in zero-shot classification by embedding input images and texts on a joint representational space. Recently, a modality gap has been reported in two-encoder contrastive models like CLIP, meaning that the image and text embeddings reside in disjoint areas of the latent space. Previous studies suggest that this gap exists due to 1) the cone effect, 2) mismatched pairs in the dataset, and 3) insufficient training. We show that, even when accounting for all these factors, and even when using the same modality, the contrastive loss actually creates a gap during training. As a result, We propose that the modality gap is inherent to the twoencoder contrastive loss and rename it the contrastive gap. We present evidence that attributes this contrastive gap to low uniformity in CLIP space, resulting in embeddings that occupy only a small portion of the latent space. To close the gap, we adapt the uniformity and alignment properties of unimodal contrastive loss to the multi-modal setting and show that simply adding these terms to the CLIP loss distributes the embeddings more uniformly in the representational space, closing the gap. In our experiments, we show that the modified representational space achieves better performance than default CLIP loss in downstream tasks such as zero-shot image classification and multi-modal arithmetic.

Fine-tuning large language models is a popular choice among users trying to adapt them for specific applications. However, fine-tuning these models is a demanding task because the user has to examine several factors, such as resource budget, runtime, model size and context length among others. A specific challenge is that fine-tuning is memory intensive, imposing constraints on the required hardware memory and context length of training data that can be handled. In this work, we share a detailed study on a variety of fine-tuning optimizations across different fine-tuning scenarios. In particular, we assess Gradient Checkpointing, Low-Rank Adaptation, DeepSpeed's Zero Redundancy Optimizer and FlashAttention. With a focus on memory and runtime, we examine the impact of different optimization combinations on GPU memory usage and execution runtime during fine-tuning phase. We provide our recommendation on the best default optimization for balancing memory and runtime across diverse model sizes. We share effective strategies for fine-tuning very large models with tens or hundreds of billions of parameters and enabling large context lengths during fine-tuning. Furthermore, we propose the appropriate optimization mixtures for fine-tuning under GPU resource limitations.