Large language models can now directly generate answers to many factual questions without referencing external sources. Unfortunately, relatively little attention has been paid to methods for evaluating the quality and correctness of these answers, for comparing the performance of one model to another, or for comparing one prompt to another. In addition, the quality of generated answers are rarely directly compared to the quality of retrieved answers. As models evolve and prompts are modified, we have no systematic way to measure improvements without resorting to expensive human judgments. To address this problem we adapt standard retrieval benchmarks to evaluate answers generated by large language models. Inspired by the BERTScore metric for summarization, we explore two approaches. In the first, we base our evaluation on the benchmark relevance judgments. We empirically run experiments on how information retrieval relevance judgments can be utilized as an anchor to evaluating the generated answers. In the second, we compare generated answers to the top results retrieved by a diverse set of retrieval models, ranging from traditional approaches to advanced methods, allowing us to measure improvements without human judgments. In both cases, we measure the similarity between an embedded representation of the generated answer and an embedded representation of a known, or assumed, relevant passage from the retrieval benchmark.
相關內容
ACM/IEEE第23屆模型驅動工程語言和系統國際會議，是模型驅動軟件和系統工程的首要會議系列，由ACM-SIGSOFT和IEEE-TCSE支持組織。自1998年以來，模型涵蓋了建模的各個方面，從語言和方法到工具和應用程序。模特的參加者來自不同的背景，包括研究人員、學者、工程師和工業專業人士。MODELS 2019是一個論壇，參與者可以圍繞建模和模型驅動的軟件和系統交流前沿研究成果和創新實踐經驗。今年的版本將為建模社區提供進一步推進建模基礎的機會，并在網絡物理系統、嵌入式系統、社會技術系統、云計算、大數據、機器學習、安全、開源等新興領域提出建模的創新應用以及可持續性。
官網鏈接： ·
echo回聲（移動應用）
·
語言模型化
·
INFORMS
·
詞元分析器
·
Recent approaches to improving the extraction of text embeddings from autoregressive large language models (LLMs) have largely focused on improvements to data, backbone pretrained language models, or improving task-differentiation via instructions. In this work, we address an architectural limitation of autoregressive models: token embeddings cannot contain information from tokens that appear later in the input. To address this limitation, we propose a simple approach, "echo embeddings," in which we repeat the input twice in context and extract embeddings from the second occurrence. We show that echo embeddings of early tokens can encode information about later tokens, allowing us to maximally leverage high-quality LLMs for embeddings. On the MTEB leaderboard, echo embeddings improve over classical embeddings by over 9% zero-shot and by around 0.7% when fine-tuned. Echo embeddings with a Mistral-7B model achieve state-of-the-art compared to prior open source models that do not leverage synthetic fine-tuning data.
Emergency communication is critical but challenging after natural disasters when ground infrastructure is devastated. Unmanned aerial vehicles (UAVs) offer enormous potential for agile relief coordination in these scenarios. However, effectively leveraging UAV fleets poses additional challenges around security, privacy, and efficient collaboration across response agencies. This paper presents a robust blockchain-enabled framework to address these challenges by integrating a consortium blockchain model, smart contracts, and cryptographic techniques to securely coordinate UAV fleets for disaster response. Specifically, we make two key contributions: a consortium blockchain architecture for secure and private multi-agency coordination; and an optimized consensus protocol balancing efficiency and fault tolerance using a delegated proof of stake practical byzantine fault tolerance (DPoS-PBFT). Comprehensive simulations showcase the framework's ability to enhance transparency, automation, scalability, and cyber-attack resilience for UAV coordination in post-disaster networks.
Contemporary language models enable new opportunities for structured reasoning with text, such as the construction and evaluation of intuitive, proof-like textual entailment trees without relying on brittle formal logic. However, progress in this direction has been hampered by a long-standing lack of a clear protocol for determining what valid compositional entailment is. This absence causes noisy datasets and limited performance gains by modern neuro-symbolic engines. To address these problems, we formulate a consistent and theoretically grounded approach to annotating decompositional entailment datasets, and evaluate its impact on LLM-based textual inference. We find that our resulting dataset, RDTE (Recognizing Decompositional Textual Entailment), has a substantially higher internal consistency (+9%) than prior decompositional entailment datasets, suggesting that RDTE is a significant step forward in the long-standing problem of forming a clear protocol for discerning entailment. We also find that training an RDTE-oriented entailment classifier via knowledge distillation and employing it in a modern neuro-symbolic reasoning engine significantly improves results (both accuracy and proof quality) over other entailment classifier baselines, illustrating the practical benefit of this advance for textual inference.
Simultaneous confidence intervals (SCIs) that are compatible with a given closed test procedure are often non-informative. More precisely, for a one-sided null hypothesis, the bound of the SCI can stick to the border of the null hypothesis, irrespective of how far the point estimate deviates from the null hypothesis. This has been illustrated for the Bonferroni-Holm and fall-back procedures, for which alternative SCIs have been suggested, that are free of this deficiency. These informative SCIs are not fully compatible with the initial multiple test, but are close to it and hence provide similar power advantages. They provide a multiple hypothesis test with strong family-wise error rate control that can be used in replacement of the initial multiple test. The current paper extends previous work for informative SCIs to graphical test procedures. The information gained from the newly suggested SCIs is shown to be always increasing with increasing evidence against a null hypothesis. The new SCIs provide a compromise between information gain and the goal to reject as many hypotheses as possible. The SCIs are defined via a family of dual graphs and the projection method. A simple iterative algorithm for the computation of the intervals is provided. A simulation study illustrates the results for a complex graphical test procedure.
We propose an adaptive ridge (AR) based estimation scheme for a heteroscedastic linear model equipped with log-linear errors. We simultaneously estimate the mean and variance parameters and show new asymptotic distributional and tightness properties in a sparse setting. We also show that estimates for zero parameters shrink with more iterations under suitable assumptions for tuning parameters. We observe possible generalizations of this paper's results through simulations and will apply the estimation method in forecasting electricity consumption.
Denoising diffusion models enable conditional generation and density modeling of complex relationships like images and text. However, the nature of the learned relationships is opaque making it difficult to understand precisely what relationships between words and parts of an image are captured, or to predict the effect of an intervention. We illuminate the fine-grained relationships learned by diffusion models by noticing a precise relationship between diffusion and information decomposition. Exact expressions for mutual information and conditional mutual information can be written in terms of the denoising model. Furthermore, pointwise estimates can be easily estimated as well, allowing us to ask questions about the relationships between specific images and captions. Decomposing information even further to understand which variables in a high-dimensional space carry information is a long-standing problem. For diffusion models, we show that a natural non-negative decomposition of mutual information emerges, allowing us to quantify informative relationships between words and pixels in an image. We exploit these new relations to measure the compositional understanding of diffusion models, to do unsupervised localization of objects in images, and to measure effects when selectively editing images through prompt interventions.
Large language models (LLMs) are pivotal in advancing natural language processing (NLP) tasks, yet their efficacy is hampered by inaccuracies and outdated knowledge. Model editing emerges as a promising solution to address these challenges. However, existing editing methods struggle to track and incorporate changes in knowledge associated with edits, which limits the generalization ability of postedit LLMs in processing edited knowledge. To tackle these problems, we propose a novel model editing method that leverages knowledge graphs for enhancing LLM editing, namely GLAME. Specifically, we first utilize a knowledge graph augmentation module to uncover associated knowledge that has changed due to editing, obtaining its internal representations within LLMs. This approach allows knowledge alterations within LLMs to be reflected through an external graph structure. Subsequently, we design a graph-based knowledge edit module to integrate structured knowledge into the model editing. This ensures that the updated parameters reflect not only the modifications of the edited knowledge but also the changes in other associated knowledge resulting from the editing process. Comprehensive experiments conducted on GPT-J and GPT-2 XL demonstrate that GLAME significantly improves the generalization capabilities of post-edit LLMs in employing edited knowledge.
Large language models (LLMs) like GPT4, have shown proficiency in generating code snippets from problem statements. Traditionally software development by humans followed a similar methodology of writing code from problem statements or requirements. However, in the past, there have been several studies that have shown the value of test-driven development (TDD) where humans write tests based on problem statements before the code for the functionality is written. In the context of LLM-based code generation, one obvious benefit of TDD is that the developer then knows for sure if the generated code has passed all the given tests or not. Therefore, in this paper, we want to empirically evaluate the hypothesis: giving the problem statements and tests as input to GPT4 is better than just giving the problem statement as input. To test our hypothesis, we build a framework TGen. In our experiments on the MBPP, HumanEval and CodeChef datasets, we consistently find that including tests solves more programming problems than not including them. Thus we show that TDD is a better development model than just using a problem statement when using GPT4 for code generation tasks.
Although large language models (LLMs) are impressive in solving various tasks, they can quickly be outdated after deployment. Maintaining their up-to-date status is a pressing concern in the current era. This paper provides a comprehensive review of recent advances in aligning LLMs with the ever-changing world knowledge without re-training from scratch. We categorize research works systemically and provide in-depth comparisons and discussion. We also discuss existing challenges and highlight future directions to facilitate research in this field. We release the paper list at //github.com/hyintell/awesome-refreshing-llms
Graph Neural Networks (GNNs) have recently been used for node and graph classification tasks with great success, but GNNs model dependencies among the attributes of nearby neighboring nodes rather than dependencies among observed node labels. In this work, we consider the task of inductive node classification using GNNs in supervised and semi-supervised settings, with the goal of incorporating label dependencies. Because current GNNs are not universal (i.e., most-expressive) graph representations, we propose a general collective learning approach to increase the representation power of any existing GNN. Our framework combines ideas from collective classification with self-supervised learning, and uses a Monte Carlo approach to sampling embeddings for inductive learning across graphs. We evaluate performance on five real-world network datasets and demonstrate consistent, significant improvement in node classification accuracy, for a variety of state-of-the-art GNNs.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191