In recent years, compression of large language models (LLMs) has emerged as an important problem to allow language model deployment on resource-constrained devices, reduce computational costs, and mitigate the environmental footprint of large-scale AI infrastructure. In this paper, we present the foundations of LLM quantization from a convex optimization perspective and propose a quantization method that builds on these foundations and outperforms previous methods. Our quantization framework, CVXQ, scales to models containing hundreds of billions of weight parameters and provides users with the flexibility to compress models to any specified model size, post-training. Large language Models (LLMs) have become a universal framework for solving a vast number of problems in natural language processing, from text translation and summarization to conversational AI and automatic generation of radiologist's reports. While LLMs outperform traditional methods in language-related tasks, they often involve tens or hundreds of billions of weight parameters, and this renders their deployment onto devices with limited resources challenging--model weights and activations no longer fit in the device memory, necessitating those activations be saved to and loaded from off-chip memory frequently during the course of inference.

Artificial Intelligence (AI) is increasingly being integrated into scientific research, particularly in the social sciences, where understanding human behavior is critical. Large Language Models (LLMs) like GPT-4 have shown promise in replicating human-like responses in various psychological experiments. However, the extent to which LLMs can effectively replace human subjects across diverse experimental contexts remains unclear. Here, we conduct a large-scale study replicating 154 psychological experiments from top social science journals with 618 main effects and 138 interaction effects using GPT-4 as a simulated participant. We find that GPT-4 successfully replicates 76.0 percent of main effects and 47.0 percent of interaction effects observed in the original studies, closely mirroring human responses in both direction and significance. However, only 19.44 percent of GPT-4's replicated confidence intervals contain the original effect sizes, with the majority of replicated effect sizes exceeding the 95 percent confidence interval of the original studies. Additionally, there is a 71.6 percent rate of unexpected significant results where the original studies reported null findings, suggesting potential overestimation or false positives. Our results demonstrate the potential of LLMs as powerful tools in psychological research but also emphasize the need for caution in interpreting AI-driven findings. While LLMs can complement human studies, they cannot yet fully replace the nuanced insights provided by human subjects.

Computer system creativity is a key step on the pathway to artificial general intelligence (AGI). It is elusive, however, due to the fact that human creativity is not fully understood and, thus, it is difficult to develop this capability in software. Large language models (LLMs) provide a facsimile of creativity and the appearance of sentience, while not actually being either creative or sentient. While LLMs have created bona fide new content, in some cases - such as with harmful hallucinations - inadvertently, their deliberate creativity is seen by some to not match that of humans. In response to this challenge, this paper proposes a technique for enhancing LLM output creativity via an iterative process of concept injection and refinement. Initial work on the development of the Creative Artificial Intelligence through Recurring Developments and Determinations (CAIRDD) system is presented and the efficacy of key system components is evaluated.

Logs play a critical role in providing essential information for system monitoring and troubleshooting. Recently, with the success of pre-trained language models (PLMs) and large language models (LLMs) in natural language processing (NLP), smaller PLMs (such as BERT) and LLMs (like ChatGPT) have become the current mainstream approaches for log analysis. While LLMs possess rich knowledge, their high computational costs and unstable performance make LLMs impractical for analyzing logs directly. In contrast, smaller PLMs can be fine-tuned for specific tasks even with limited computational resources, making them more practical. However, these smaller PLMs face challenges in understanding logs comprehensively due to their limited expert knowledge. To better utilize the knowledge embedded within LLMs for log understanding, this paper introduces a novel knowledge enhancement framework, called LUK, which acquires expert knowledge from LLMs to empower log understanding on a smaller PLM. Specifically, we design a multi-expert collaboration framework based on LLMs consisting of different roles to acquire expert knowledge. In addition, we propose two novel pre-training tasks to enhance the log pre-training with expert knowledge. LUK achieves state-of-the-art results on different log analysis tasks and extensive experiments demonstrate expert knowledge from LLMs can be utilized more effectively to understand logs.

In this work, we explore the use and reliability of Large Language Models (LLMs) in musicology. From a discussion with experts and students, we assess the current acceptance and concerns regarding this, nowadays ubiquitous, technology. We aim to go one step further, proposing a semi-automatic method to create an initial benchmark using retrieval-augmented generation models and multiple-choice question generation, validated by human experts. Our evaluation on 400 human-validated questions shows that current vanilla LLMs are less reliable than retrieval augmented generation from music dictionaries. This paper suggests that the potential of LLMs in musicology requires musicology driven research that can specialized LLMs by including accurate and reliable domain knowledge.

Overcoming the limited context limitations in early-generation LLMs, retrieval-augmented generation (RAG) has been a reliable solution for context-based answer generation in the past. Recently, the emergence of long-context LLMs allows the models to incorporate much longer text sequences, making RAG less attractive. Recent studies show that long-context LLMs significantly outperform RAG in long-context applications. Unlike the existing works favoring the long-context LLM over RAG, we argue that the extremely long context in LLMs suffers from a diminished focus on relevant information and leads to potential degradation in answer quality. This paper revisits the RAG in long-context answer generation. We propose an order-preserve retrieval-augmented generation (OP-RAG) mechanism, which significantly improves the performance of RAG for long-context question-answer applications. With OP-RAG, as the number of retrieved chunks increases, the answer quality initially rises, and then declines, forming an inverted U-shaped curve. There exist sweet points where OP-RAG could achieve higher answer quality with much less tokens than long-context LLM taking the whole context as input. Extensive experiments on public benchmark demonstrate the superiority of our OP-RAG.

This study explores the potential of large language models (LLMs) for identifying and examining intertextual relationships within biblical, Koine Greek texts. By evaluating the performance of LLMs on various intertextuality scenarios the study demonstrates that these models can detect direct quotations, allusions, and echoes between texts. The LLM's ability to generate novel intertextual observations and connections highlights its potential to uncover new insights. However, the model also struggles with long query passages and the inclusion of false intertextual dependences, emphasizing the importance of expert evaluation. The expert-in-the-loop methodology presented offers a scalable approach for intertextual research into the complex web of intertextuality within and beyond the biblical corpus.

When language models (LMs) are trained to forget (or "unlearn'') a skill, how precisely does their behavior change? We study the behavior of transformer LMs in which tasks have been forgotten via fine-tuning on randomized labels. Such LMs learn to generate near-random predictions for individual examples in the "training'' set used for forgetting. Across tasks, however, LMs exhibit extreme variability in whether LM predictions change on examples outside the training set. In some tasks (like entailment classification), forgetting generalizes robustly, and causes models to produce uninformative predictions on new task instances; in other tasks (like physical commonsense reasoning and scientific question answering) forgetting affects only the training examples, and models continue to perform the "forgotten'' task accurately even for examples very similar to those that appeared in the training set. Dataset difficulty is not predictive of whether a behavior can be forgotten; instead, generalization in forgetting is (weakly) predicted by the confidence of LMs' initial task predictions and the variability of LM representations of training data, with low confidence and low variability both associated with greater generalization. Perhaps most surprisingly, random-label forgetting appears to be somewhat insensitive to the contents of the training set: for example, models trained on science questions with random labels continue to answer other science questions accurately, but begin to produce random labels on entailment classification tasks. Finally, we show that even generalizable forgetting is shallow: linear probes trained on LMs' representations can still perform tasks reliably after forgetting. Our results highlight the difficulty and unpredictability of performing targeted skill removal from models via fine-tuning.

This paper presents rerankers, a Python library which provides an easy-to-use interface to the most commonly used re-ranking approaches. Re-ranking is an integral component of many retrieval pipelines; however, there exist numerous approaches to it, relying on different implementation methods. rerankers unifies these methods into a single user-friendly interface, allowing practitioners and researchers alike to explore different methods while only changing a single line of Python code. Moreover ,rerankers ensures that its implementations are done with the fewest dependencies possible, and re-uses the original implementation whenever possible, guaranteeing that our simplified interface results in no performance degradation compared to more complex ones. The full source code and list of supported models are updated regularly and available at https://github.com/answerdotai/rerankers.

The recent advances in large language models (LLMs) have significantly expanded their applications across various fields such as language generation, summarization, and complex question answering. However, their application to privacy compliance and technical privacy reviews remains under-explored, raising critical concerns about their ability to adhere to global privacy standards and protect sensitive user data. This paper seeks to address this gap by providing a comprehensive case study evaluating LLMs' performance in privacy-related tasks such as privacy information extraction (PIE), legal and regulatory key point detection (KPD), and question answering (QA) with respect to privacy policies and data protection regulations. We introduce a Privacy Technical Review (PTR) framework, highlighting its role in mitigating privacy risks during the software development life-cycle. Through an empirical assessment, we investigate the capacity of several prominent LLMs, including BERT, GPT-3.5, GPT-4, and custom models, in executing privacy compliance checks and technical privacy reviews. Our experiments benchmark the models across multiple dimensions, focusing on their precision, recall, and F1-scores in extracting privacy-sensitive information and detecting key regulatory compliance points. While LLMs show promise in automating privacy reviews and identifying regulatory discrepancies, significant gaps persist in their ability to fully comply with evolving legal standards. We provide actionable recommendations for enhancing LLMs' capabilities in privacy compliance, emphasizing the need for robust model improvements and better integration with legal and regulatory requirements. This study underscores the growing importance of developing privacy-aware LLMs that can both support businesses in compliance efforts and safeguard user privacy rights.