Modern ML systems increasingly augment input instances with additional relevant information to enhance final prediction. Despite growing interest in such retrieval-augmented models, their fundamental properties and training are not well understood. We propose a statistical framework to study such models with two components: 1) a {\em retriever} to identify the relevant information out of a large corpus via a data-dependent metric; and 2) a {\em predictor} that consumes the input instances along with the retrieved information to make the final predictions. We present a principled method for end-to-end training of both components and draw connections with various training approaches in the literature. Furthermore, we establish excess risk bounds for retrieval-augmented models while delineating the contributions of both retriever and predictor towards the model performance. We validate the utility of our proposed training methods along with the key takeaways from our statistical analysis on open domain question answering task where retrieval augmentation is important.

Verifiers or reward models are often used to enhance the reasoning performance of large language models (LLMs). A common approach is the Best-of-N method, where N candidate solutions generated by the LLM are ranked by a verifier, and the best one is selected. While LLM-based verifiers are typically trained as discriminative classifiers to score solutions, they do not utilize the text generation capabilities of pretrained LLMs. To overcome this limitation, we instead propose training verifiers using the ubiquitous next-token prediction objective, jointly on verification and solution generation. Compared to standard verifiers, such generative verifiers (GenRM) can benefit from several advantages of LLMs: they integrate seamlessly with instruction tuning, enable chain-of-thought reasoning, and can utilize additional inference-time compute via majority voting for better verification. We demonstrate that when using Gemma-based verifiers on algorithmic and grade-school math reasoning tasks, GenRM outperforms discriminative verifiers and LLM-as-a-Judge, showing a 16-64% improvement in the percentage of problems solved with Best-of-N. Furthermore, we show that GenRM scales favorably across dataset size, model capacity, and inference-time compute.

This research introduces the Multilevel Embedding Association Test (ML-EAT), a method designed for interpretable and transparent measurement of intrinsic bias in language technologies. The ML-EAT addresses issues of ambiguity and difficulty in interpreting the traditional EAT measurement by quantifying bias at three levels of increasing granularity: the differential association between two target concepts with two attribute concepts; the individual effect size of each target concept with two attribute concepts; and the association between each individual target concept and each individual attribute concept. Using the ML-EAT, this research defines a taxonomy of EAT patterns describing the nine possible outcomes of an embedding association test, each of which is associated with a unique EAT-Map, a novel four-quadrant visualization for interpreting the ML-EAT. Empirical analysis of static and diachronic word embeddings, GPT-2 language models, and a CLIP language-and-image model shows that EAT patterns add otherwise unobservable information about the component biases that make up an EAT; reveal the effects of prompting in zero-shot models; and can also identify situations when cosine similarity is an ineffective metric, rendering an EAT unreliable. Our work contributes a method for rendering bias more observable and interpretable, improving the transparency of computational investigations into human minds and societies.

Large language models (LLMs) have shown high agreement with human raters across a variety of tasks, demonstrating potential to ease the challenges of human data collection. In computational social science (CSS), researchers are increasingly leveraging LLM annotations to complement slow and expensive human annotations. Still, guidelines for collecting and using LLM annotations, without compromising the validity of downstream conclusions, remain limited. We introduce Confidence-Driven Inference: a method that combines LLM annotations and LLM confidence indicators to strategically select which human annotations should be collected, with the goal of producing accurate statistical estimates and provably valid confidence intervals while reducing the number of human annotations needed. Our approach comes with safeguards against LLM annotations of poor quality, guaranteeing that the conclusions will be both valid and no less accurate than if we only relied on human annotations. We demonstrate the effectiveness of Confidence-Driven Inference over baselines in statistical estimation tasks across three CSS settings--text politeness, stance, and bias--reducing the needed number of human annotations by over 25% in each. Although we use CSS settings for demonstration, Confidence-Driven Inference can be used to estimate most standard quantities across a broad range of NLP problems.

Semiconductors, crucial to modern electronics, are generally under-researched in foundational models. It highlights the need for research to enhance the semiconductor device technology portfolio and aid in high-end device fabrication. In this paper, we introduce sLAVA, a small-scale vision-language assistant tailored for semiconductor manufacturing, with a focus on electron microscopy image analysis. It addresses challenges of data scarcity and acquiring high-quality, expert-annotated data. We employ a teacher-student paradigm, using a foundational vision language model like GPT-4 as a teacher to create instruction-following multimodal data for customizing the student model, sLAVA, for electron microscopic image analysis tasks on consumer hardware with limited budgets. Our approach allows enterprises to further fine-tune the proposed framework with their proprietary data securely within their own infrastructure, protecting intellectual property. Rigorous experiments validate that our framework surpasses traditional methods, handles data shifts, and enables high-throughput screening.

Parameter Efficient Fine-Tuning (PEFT) methods have gained popularity and democratized the usage of Large Language Models (LLMs). Recent studies have shown that a small subset of weights significantly impacts performance. Based on this observation, we introduce a novel PEFT method, called Gaussian noise Injected Fine Tuning of Salient Weights (GIFT-SW). Our method updates only salient columns, while injecting Gaussian noise into non-salient ones. To identify these columns, we developeda generalized sensitivity metric that extends and unifies metrics from previous studies. Experiments with LLaMA models demonstrate that GIFT-SW outperforms full fine-tuning and modern PEFT methods under the same computational budget. Moreover, GIFT-SW offers practical advantages to recover performance of models subjected to mixed-precision quantization with keeping salient weights in full precision.

Supervised learning-based software vulnerability detectors often fall short due to the inadequate availability of labelled training data. In contrast, Large Language Models (LLMs) such as GPT-4, are not trained on labelled data, but when prompted to detect vulnerabilities, LLM prediction accuracy is only marginally better than random guessing. In this paper, we explore a different approach by reframing vulnerability detection as one of anomaly detection. Since the vast majority of code does not contain vulnerabilities and LLMs are trained on massive amounts of such code, vulnerable code can be viewed as an anomaly from the LLM's predicted code distribution, freeing the model from the need for labelled data to provide a learnable representation of vulnerable code. Leveraging this perspective, we demonstrate that LLMs trained for code generation exhibit a significant gap in prediction accuracy when prompted to reconstruct vulnerable versus non-vulnerable code. Using this insight, we implement ANVIL, a detector that identifies software vulnerabilities at line-level granularity. Our experiments explore the discriminating power of different anomaly scoring methods, as well as the sensitivity of ANVIL to context size. We also study the effectiveness of ANVIL on various LLM families, and conduct leakage experiments on vulnerabilities that were discovered after the knowledge cutoff of our evaluated LLMs. On a collection of vulnerabilities from the Magma benchmark, ANVIL outperforms state-of-the-art line-level vulnerability detectors, LineVul and LineVD, which have been trained with labelled data, despite ANVIL having never been trained with labelled vulnerabilities. Specifically, our approach achieves $1.62\times$ to $2.18\times$ better Top-5 accuracies and $1.02\times$ to $1.29\times$ times better ROC scores on line-level vulnerability detection tasks.

We have observed a distinctive quantization-related behavior in the LLaMA3/3.1-70B models that is absent in both the LLaMA2-70B and LLaMA3/3.1-8B/405B models. Quantization is a crucial technique for deploying large language models (LLMs) efficiently. Among various bit widths and representations for weights and activations, the 8-bit integer weight and 8-bit integer activation (W8A8) configuration is particularly popular due to its widespread hardware support. However, the impact of W8A8 post-training quantization on model accuracy remains contentious. While several studies have suggested calibrating either weights or activations to mitigate accuracy degradation, a comprehensive solution has yet to be identified. In this paper, we empirically investigate multiple LLMs featured on an open LLM leaderboard, discovering that the LLaMA3-70B model series have a unique accuracy degradation behavior with W8A8 per-channel post-training quantization. In contrast, other model series such as LLaMA2, LLaMA3-8B, Qwen, Mixtral, Mistral, Phi-3, and Falcon demonstrate robust performance with W8A8, sometimes surpassing their FP16 counterparts. Contrary to previous assertions attributing degradation to the large dynamic range of activations, our findings indicate that the weight distribution of the LLaMA3-70B is the primary factor behind the vulnerability. By meticulously analyzing the distinct characteristics of weight distributions across Transformer blocks, we propose a mixed strategy with less than 3% of the layers enabling finer W8A8 quantization granularity, while the remaining 97% of layers retain the per-channel configuration. As a result, the average accuracy of LLaMA3-70B-W8A8 is increased from 45.5% to 73.4% (just 0.7% shy of LLaMA3-70B-FP16) across eight reasoning tasks. Notably, our method requires neither calibration nor fine-tuning.

In recent years, the rapid expansion of model sizes has led to large-scale pre-trained models demonstrating remarkable capabilities. Consequently, there has been a trend towards increasing the scale of models. However, this trend introduces significant challenges, including substantial computational costs of training and transfer to downstream tasks. To address these issues, Parameter-Efficient Fine-Tuning (PEFT) methods have been introduced. These methods optimize large-scale pre-trained models for specific tasks by fine-tuning a select group of parameters. Among these PEFT methods, adapter-based and prompt-based methods are the primary techniques. Specifically, in the field of visual fine-tuning, adapters gain prominence over prompts because of the latter's relatively weaker performance and efficiency. Under the circumstances, we refine the widely-used Visual Prompt Tuning (VPT) method, proposing Cross Visual Prompt Tuning (CVPT). CVPT calculates cross-attention between the prompt tokens and the embedded tokens, which allows us to compute the semantic relationship between them and conduct the fine-tuning of models exactly to adapt visual tasks better. Furthermore, we introduce the weight-sharing mechanism to initialize the parameters of cross-attention, which avoids massive learnable parameters from cross-attention and enhances the representative capability of cross-attention. We conduct comprehensive testing across 25 datasets and the result indicates that CVPT significantly improves VPT's performance and efficiency in visual tasks. For example, on the VTAB-1K benchmark, CVPT outperforms VPT over 4% in average accuracy, rivaling the advanced adapter-based methods in performance and efficiency. Our experiments confirm that prompt-based methods can achieve exceptional results in visual fine-tuning.

Large Language Models (LLMs) have gained significant attention due to their high performance on a wide range of natural language tasks since the release of ChatGPT. The LLMs learn to understand and generate language by training billions of model parameters on vast volumes of text data. Despite being a relatively new field, LLM research is rapidly advancing in various directions. In this paper, we present an overview of LLM families, including LLaMA, PaLM, GPT, and MoE, and the methods developed to create and enhance LLMs for official European Union (EU) languages. We provide a comprehensive summary of common monolingual and multilingual datasets used for pretraining large language models.