We introduce and study the problem of adversarial arithmetic, which provides a simple yet challenging testbed for language model alignment. This problem is comprised of arithmetic questions posed in natural language, with an arbitrary adversarial string inserted before the question is complete. Even in the simple setting of 1-digit addition problems, it is easy to find adversarial prompts that make all tested models (including PaLM2, GPT4, Claude2) misbehave, and even to steer models to a particular wrong answer. We additionally provide a simple algorithm for finding successful attacks by querying those same models, which we name "prompt inversion rejection sampling" (PIRS). We finally show that models can be partially hardened against these attacks via reinforcement learning and via agentic constitutional loops. However, we were not able to make a language model fully robust against adversarial arithmetic attacks.

Few-shot relation extraction involves identifying the type of relationship between two specific entities within a text, using a limited number of annotated samples. A variety of solutions to this problem have emerged by applying meta-learning and neural graph techniques which typically necessitate a training process for adaptation. Recently, the strategy of in-context learning has been demonstrating notable results without the need of training. Few studies have already utilized in-context learning for zero-shot information extraction. Unfortunately, the evidence for inference is either not considered or implicitly modeled during the construction of chain-of-thought prompts. In this paper, we propose a novel approach for few-shot relation extraction using large language models, named CoT-ER, chain-of-thought with explicit evidence reasoning. In particular, CoT-ER first induces large language models to generate evidences using task-specific and concept-level knowledge. Then these evidences are explicitly incorporated into chain-of-thought prompting for relation extraction. Experimental results demonstrate that our CoT-ER approach (with 0% training data) achieves competitive performance compared to the fully-supervised (with 100% training data) state-of-the-art approach on the FewRel1.0 and FewRel2.0 datasets.

An ideal length-extrapolatable Transformer language model can handle sequences longer than the training length without any fine-tuning. Such long-context utilization capability relies heavily on a flexible positional embedding design. Upon investigating the flexibility of existing large pre-trained Transformer language models, we find that the T5 family deserves a closer look, as its positional embeddings capture rich and flexible attention patterns. However, T5 suffers from the dispersed attention issue: the longer the input sequence, the flatter the attention distribution. To alleviate the issue, we propose two attention alignment strategies via temperature scaling. Our findings show improvement on the long-context utilization capability of T5 on language modeling, retrieval, multi-document question answering, and code completion tasks without any fine-tuning. This suggests that a flexible positional embedding design and attention alignment can go a long way toward Transformer length extrapolation.

Recent advances have led to the availability of many pre-trained language models (PLMs); however, a question that remains is how much data is truly needed to fine-tune PLMs for downstream tasks? In this work, we introduce DEFT, a data-efficient fine-tuning framework that leverages unsupervised core-set selection to minimize the amount of data needed to fine-tune PLMs for downstream tasks. We demonstrate the efficacy of our DEFT framework in the context of text-editing LMs, and compare to the state-of-the art text-editing model, CoEDIT. Our quantitative and qualitative results demonstrate that DEFT models are just as accurate as CoEDIT while being finetuned on ~70% less data.

Large language models (LLMs) are now available in various sizes and configurations from cloud API providers. While this diversity offers a broad spectrum of choices, effectively leveraging the options to optimize computational cost and performance remains challenging. In this work, we present AutoMix, an approach that strategically routes queries to larger LMs, based on the approximate correctness of outputs from a smaller LM. Central to AutoMix is a few-shot self-verification mechanism, which estimates the reliability of its own outputs without requiring training. Given that verifications can be noisy, we employ a meta verifier in AutoMix to refine the accuracy of these assessments. Our experiments using LLAMA2-13/70B, on five context-grounded reasoning datasets demonstrate that AutoMix surpasses established baselines, improving the incremental benefit per cost by up to 89%. Our code and data are available at https://github.com/automix-llm/automix.

Recent studies show that traditional fairytales are rife with harmful gender biases. To help mitigate these gender biases in fairytales, this work aims to assess learned biases of language models by evaluating their robustness against gender perturbations. Specifically, we focus on Question Answering (QA) tasks in fairytales. Using counterfactual data augmentation to the FairytaleQA dataset, we evaluate model robustness against swapped gender character information, and then mitigate learned biases by introducing counterfactual gender stereotypes during training time. We additionally introduce a novel approach that utilizes the massive vocabulary of language models to support text genres beyond fairytales. Our experimental results suggest that models are sensitive to gender perturbations, with significant performance drops compared to the original testing set. However, when first fine-tuned on a counterfactual training dataset, models are less sensitive to the later introduced anti-gender stereotyped text.

Zero-shot cross-lingual generation assumes finetuning the multilingual pretrained language model (mPLM) on a generation task in one language and then using it to make predictions for this task in other languages. Previous works notice a frequent problem of generation in a wrong language and propose approaches to address it, usually using mT5 as a backbone model. In this work, we test alternative mPLMs, such as mBART and NLLB-200, and compare various approaches proposed in the literature in a unified setting. We first underline the importance of tuning learning rate used for finetuning, which helps to substantially alleviate the problem of generation in the wrong language. Then, we show that with careful learning rate tuning, the simple full finetuning of the model acts as a very strong baseline; other competitive approaches include parameter-efficient tuning with adapters and training on several source languages. Finally, we find that mBART performs similarly to mT5 of the same size, and NLLB-200 can be competitive in some cases.

In this paper, we address the concept of "alignment" in large language models (LLMs) through the lens of post-structuralist socio-political theory, specifically examining its parallels to empty signifiers. To establish a shared vocabulary around how abstract concepts of alignment are operationalised in empirical datasets, we propose a framework that demarcates: 1) which dimensions of model behaviour are considered important, then 2) how meanings and definitions are ascribed to these dimensions, and by whom. We situate existing empirical literature and provide guidance on deciding which paradigm to follow. Through this framework, we aim to foster a culture of transparency and critical evaluation, aiding the community in navigating the complexities of aligning LLMs with human populations.

Static analysis is a widely used technique in software engineering for identifying and mitigating bugs. However, a significant hurdle lies in achieving a delicate balance between precision and scalability. Large Language Models (LLMs) offer a promising alternative, as recent advances demonstrate remarkable capabilities in comprehending, generating, and even debugging code. Yet, the logic of bugs can be complex and require sophisticated reasoning and a large analysis scope spanning multiple functions. Therefore, at this point, LLMs are better used in an assistive role to complement static analysis. In this paper, we take a deep dive into the open space of LLM-assisted static analysis, using use-before-initialization (UBI) bugs as a case study. To this end, we develop LLift, a fully automated framework that interfaces with both a static analysis tool and an LLM. By carefully designing the framework and the prompts, we are able to overcome a number of challenges, including bug-specific modeling, the large problem scope, the non-deterministic nature of LLMs, etc. Tested in a real-world scenario analyzing nearly a thousand potential UBI bugs produced by static analysis, LLift demonstrates a potent capability, showcasing a reasonable precision (50%) and appearing to have no missing bugs. It even identified 13 previously unknown UBI bugs in the Linux kernel. This research paves the way for new opportunities and methodologies in using LLMs for bug discovery in extensive, real-world datasets.

Parameter-efficient tuning (PET) has been widely explored in recent years because it tunes much fewer parameters (PET modules) than full-parameter fine-tuning (FT) while still stimulating sufficient knowledge from large language models (LLMs) for downstream tasks. Moreover, when PET is employed to serve multiple tasks, different task-specific PET modules can be built on a frozen LLM, avoiding redundant LLM deployments. Although PET significantly reduces the cost of tuning and deploying LLMs, its inference still suffers from the computational bottleneck of LLMs. To address the above issue, we propose an effective PET framework based on compressed LLMs, named "CPET". In CPET, we evaluate the impact of mainstream LLM compression techniques on PET performance and then introduce knowledge inheritance and recovery strategies to restore the knowledge loss caused by these compression techniques. Our experimental results demonstrate that, owing to the restoring strategies of CPET, collaborating task-specific PET modules with a compressed LLM can achieve comparable performance to collaborating PET modules with the original version of the compressed LLM and outperform directly applying vanilla PET methods to the compressed LLM.