Machine Learning (ML) in low-data settings remains an underappreciated yet crucial problem. Hence, data augmentation methods to increase the sample size of datasets needed for ML are key to unlocking the transformative potential of ML in data-deprived regions and domains. Unfortunately, the limited training set constrains traditional tabular synthetic data generators in their ability to generate a large and diverse augmented dataset needed for ML tasks. To address this challenge, we introduce CLLM, which leverages the prior knowledge of Large Language Models (LLMs) for data augmentation in the low-data regime. However, not all the data generated by LLMs will improve downstream utility, as for any generative model. Consequently, we introduce a principled curation mechanism, leveraging learning dynamics, coupled with confidence and uncertainty metrics, to obtain a high-quality dataset. Empirically, on multiple real-world datasets, we demonstrate the superior performance of CLLM in the low-data regime compared to conventional generators. Additionally, we provide insights into the LLM generation and curation mechanism, shedding light on the features that enable them to output high-quality augmented datasets.

Protecting the copyright of large language models (LLMs) has become crucial due to their resource-intensive training and accompanying carefully designed licenses. However, identifying the original base model of an LLM is challenging due to potential parameter alterations. In this study, we introduce a human-readable fingerprint for LLMs that uniquely identifies the base model without exposing model parameters or interfering with training. We first observe that the vector direction of LLM parameters remains stable after the model has converged during pretraining, showing negligible perturbations through subsequent training steps, including continued pretraining, supervised fine-tuning (SFT), and RLHF, which makes it a sufficient condition to identify the base model. The necessity is validated by continuing to train an LLM with an extra term to drive away the model parameters' direction and the model becomes damaged. However, this direction is vulnerable to simple attacks like dimension permutation or matrix rotation, which significantly change it without affecting performance. To address this, leveraging the Transformer structure, we systematically analyze potential attacks and define three invariant terms that identify an LLM's base model. We make these invariant terms human-readable by mapping them to a Gaussian vector using a convolutional encoder and then converting it into a natural image with StyleGAN2. Our method generates a dog image as an identity fingerprint for an LLM, where the dog's appearance strongly indicates the LLM's base model. The fingerprint provides intuitive information for qualitative discrimination, while the invariant terms can be employed for quantitative and precise verification. Experimental results across various LLMs demonstrate the effectiveness of our method.

Large language models (LLMs) have demonstrated their potential in social science research by emulating human perceptions and behaviors, a concept referred to as algorithmic fidelity. The LLMs were conditioned on demographics and/or psychological covariates to simulate survey responses. The findings indicate that LLMs can effectively capture presidential voting behaviors but encounter challenges in accurately representing global warming perspectives when relevant covariates are not included. GPT-4 exhibits improved performance when conditioned on both demographics and covariates. However, disparities emerge in LLM estimations of the views of certain groups, with LLMs tending to underestimate worry about global warming among Black Americans. While highlighting the potential of LLMs to aid social science research, these results underscore the importance of meticulous conditioning, model selection, survey question format, and bias assessment when employing LLMs for survey simulation. Further investigation into prompt engineering and algorithm auditing is essential to harness the power of LLMs while addressing their inherent limitations. Keywords: Global warming; large language models; algorithmic fidelity; public opinion 1. Introduction It is very important to measure public opinion about global warming, as these opinions can have considerable influence over policy-making decisions (Bromley-Trujillo & Poe, 2020) and shape public behavior (Doherty & Webler, 2016). A primary method employed by scholars and policymakers for measuring and assessing these opinions is through representative surveys (Berinsky, 2017). However, the extensive time and financial resources required for these surveys can hinder the timely tracking of evolving public opinions about global warming. Resource constraints can also lead to an unintended bias towards majority opinions, potentially neglecting the perspectives of minority groups due to their typically smaller sample sizes in national representative surveys. Nonetheless, understanding diverse public opinion regarding global warming is also vital for climate justice. This understanding can promote equitable decisionmaking, elevate the concerns of vulnerable communities, help align climate policies with democratic principles, build public support, and address disparities in climate change awareness and priorities. Furthermore, understanding the diversity of public opinion can help support a just transition and mobilize support for climate justice initiatives.

Space-based gravitational wave detection is one of the most anticipated gravitational wave (GW) detection projects in the next decade, which is promising to detect abundant compact binary systems. However, the precise prediction of space GW waveforms remains unexplored. To solve the data processing difficulty in the increasing waveform complexity caused by detectors' response and second-generation time-delay interferometry (TDI 2.0), an interpretable pre-trained large model named CBS-GPT (Compact Binary Systems Waveform Generation with Generative Pre-trained Transformer) is proposed. For compact binary system waveforms, three models were trained to predict the waveforms of massive black hole binary (MBHB), extreme mass-ratio inspirals (EMRIs), and galactic binary (GB), achieving prediction accuracies of 99%, 91%, and 99%, respectively at most.The CBS-GPT model exhibits notable generalization and interpretability, with its hidden parameters effectively capturing the intricate information of waveforms, even with complex instrument response and a wide parameter range. Our research demonstrates the potential of large pre-trained models in gravitational wave realm, opening up new opportunities and guidance for future researches such as the complex waveforms generation, gap completion, and deep learning model design for GW science.

Automated fact-checking, using machine learning to verify claims, has grown vital as misinformation spreads beyond human fact-checking capacity. Large Language Models (LLMs) like GPT-4 are increasingly trusted to write academic papers, lawsuits, and news articles and to verify information, emphasizing their role in discerning truth from falsehood and the importance of being able to verify their outputs. Understanding the capacities and limitations of LLMs in fact-checking tasks is therefore essential for ensuring the health of our information ecosystem. Here, we evaluate the use of LLM agents in fact-checking by having them phrase queries, retrieve contextual data, and make decisions. Importantly, in our framework, agents explain their reasoning and cite the relevant sources from the retrieved context. Our results show the enhanced prowess of LLMs when equipped with contextual information. GPT-4 outperforms GPT-3, but accuracy varies based on query language and claim veracity. While LLMs show promise in fact-checking, caution is essential due to inconsistent accuracy. Our investigation calls for further research, fostering a deeper comprehension of when agents succeed and when they fail.

The proliferation of large AI models trained on uncurated, often sensitive web-scraped data has raised significant privacy concerns. One of the concerns is that adversaries can extract information about the training data using privacy attacks. Unfortunately, the task of removing specific information from the models without sacrificing performance is not straightforward and has proven to be challenging. We propose a rather easy yet effective defense based on backdoor attacks to remove private information such as names and faces of individuals from vision-language models by fine-tuning them for only a few minutes instead of re-training them from scratch. Specifically, through strategic insertion of backdoors into text encoders, we align the embeddings of sensitive phrases with those of neutral terms-"a person" instead of the person's actual name. For image encoders, we map embeddings of individuals to be removed from the model to a universal, anonymous embedding. Our empirical results demonstrate the effectiveness of our backdoor-based defense on CLIP by assessing its performance using a specialized privacy attack for zero-shot classifiers. Our approach provides not only a new "dual-use" perspective on backdoor attacks, but also presents a promising avenue to enhance the privacy of individuals within models trained on uncurated web-scraped data.

Large language models (LLMs) fine-tuned by reinforcement learning from human feedback (RLHF) are becoming more widely deployed. We coin the term $\textit{Implicit Reward Model}$ (IRM) to refer to the changes that occur to an LLM during RLHF that result in high-reward generations. We interpret IRMs, and measure their divergence from the RLHF reward model used in the fine-tuning process that induced them. By fitting a linear function to an LLM's IRM, a reward model with the same type signature as the RLHF reward model is constructed, allowing for direct comparison. Additionally, we validate our construction of the IRM through cross-comparison with classifications of features generated by an LLM based on their relevance to the RLHF reward model. Better comprehending IRMs can help minimize discrepencies between LLM behavior and training objectives, which we believe to be an essential component of the $\textit{safety}$ and $\textit{alignment}$ of LLMs.

We present an in-context learning agent for formal theorem-proving in environments like Lean and Coq. Current state-of-the-art models for the problem are finetuned on environment-specific proof data. By contrast, our approach, called COPRA, repeatedly asks a high-capacity, general-purpose large language model (GPT-4) to propose tactic applications from within a stateful backtracking search. Proposed tactics are executed in the underlying proof environment. Feedback from the execution is used to build the prompt for the next model query, along with selected information from the search history and lemmas retrieved from an external database. We evaluate our implementation of COPRA on the miniF2F benchmark for Lean and a set of Coq tasks from the CompCert project. On these benchmarks, COPRA significantly outperforms few-shot invocations of GPT-4. It also compares favorably against finetuning-based approaches, outperforming REPROVER, a state-of-the-art finetuned approach for Lean, in terms of the pass@1 metric. Our code and data are available at https://github.com/trishullab/copra.

Large language models (LLMs) exhibit remarkable performance improvement through in-context learning (ICL) by leveraging task-specific examples in the input. However, the mechanisms behind this improvement remain elusive. In this work, we investigate how LLM embeddings and attention representations change following in-context-learning, and how these changes mediate improvement in behavior. We employ neuroscience-inspired techniques such as representational similarity analysis (RSA) and propose novel methods for parameterized probing and measuring ratio of attention to relevant vs. irrelevant information in Llama-2 70B and Vicuna 13B. We designed two tasks with a priori relationships among their conditions: linear regression and reading comprehension. We formed hypotheses about expected similarities in task representations and measured hypothesis alignment of LLM representations before and after ICL as well as changes in attention. Our analyses revealed a meaningful correlation between improvements in behavior after ICL and changes in both embeddings and attention weights across LLM layers. This empirical framework empowers a nuanced understanding of how latent representations shape LLM behavior, offering valuable tools and insights for future research and practical applications.

The presenter pasted in what he called "about 16 pages' worth of tax code" These seven sentences about Alice, Bob, and Charlie come word-for-word from a handcrafted data set we developed at Johns Hopkins University and published in 2020 for training and measuring AI models for reasoning over statutory language. Every word, punctuation mark, and Maryland; Nils number in the taxpayer facts comes exactly from Holzenberger is an our tax_case_9 -- even the percent sign at the start associate professor in of the line. This work has been supported by the U.S. National Science Foundation under grant No. 2204926. The entire livestream is available at OpenAI, "GPT-4 Developer The tax law example starts at minute 19:11. Go to the directory "Cases" to find the file tax_case_9.pl. Tax_case_9.pl is written in the programming language Prolog. Federal content, please visit www.taxnotes.com. Where did the "about 16 pages' worth of tax out the TCJA standard deduction increase at code" come from? Again, from our 2020 data set. SARA has two deduction for 2018 was $24,000. From minute 20:07 to 20:40 of the livestream, handcrafted cases in SARA; tax_case_9 is one of we see some of the tax sections pasted into GPT-4. The statutes consist of nine sections of the These are SARA's heavily edited version of the IRC, For example, at and remove ambiguity. If you put all the SARA 20:23, we see part of section 63(c) with the statutes into a single file it will be about 16 pages paragraphs jumping from (3) to (5); in SARA, we long (depending on the font). At 20:26, we see part of section One of our edits was paring section 1 down to 63(c)(6) with only subparagraphs (A), (B), and (D); only sections 1(a) through (d), which contain the in SARA, we edited out (C). At 20:40, we see parts Clinton-era tax rates. We cut section 1(j), which of section 3306(b) with the paragraphs jumping contains the reduced Tax Cuts and Jobs Act rates from (2) to (7); in SARA, we edited out paragraphs for 2018-2025. This editing explains why GPT-4 (3) through (6). At 20:39 we see sections 3301 and got the wrong answer on the livestream for Alice 3306 regarding the federal unemployment tax; and Bob's 2018 taxes. We did not, however, edit while these two sections are irrelevant to Alice and Bob's tax liability in tax_case_9, they are two The author Holzenberger did all the handcrafting and hand editing. Federal content, please visit www.taxnotes.com. You can We empirically verified that using the SARA download our data set and compare it with the version of the IRC causes GPT-4 to get the wrong livestream's recording on YouTube. First, we The presenter then gives directions to GPT-4: pasted into GPT-4 all nine SARA statutes, plus our "Now calculate their total liability." GPT-4 gives facts about Alice, Bob, and Charlie. Then we detailed step-by-step calculations and concludes used the same "Now calculate their total liability" that "Alice and Bob's total tax liability for 2018 is command.