Transfer learning is important for foundation models to adapt to downstream tasks. However, many foundation models are proprietary, so users must share their data with model owners to fine-tune the models, which is costly and raise privacy concerns. Moreover, fine-tuning large foundation models is computation-intensive and impractical for most downstream users. In this paper, we propose Offsite-Tuning, a privacy-preserving and efficient transfer learning framework that can adapt billion-parameter foundation models to downstream data without access to the full model. In offsite-tuning, the model owner sends a light-weight adapter and a lossy compressed emulator to the data owner, who then fine-tunes the adapter on the downstream data with the emulator's assistance. The fine-tuned adapter is then returned to the model owner, who plugs it into the full model to create an adapted foundation model. Offsite-tuning preserves both parties' privacy and is computationally more efficient than the existing fine-tuning methods that require access to the full model weights. We demonstrate the effectiveness of offsite-tuning on various large language and vision foundation models. Offsite-tuning can achieve comparable accuracy as full model fine-tuning while being privacy-preserving and efficient, achieving 6.5x speedup and 5.6x memory reduction. Code is available at https://github.com/mit-han-lab/offsite-tuning.

We provide a new efficient version of the backpropagation algorithm, specialized to the case where the weights of the neural network being trained are sparse. Our algorithm is general, as it applies to arbitrary (unstructured) sparsity and common layer types (e.g., convolutional or linear). We provide a fast vectorized implementation on commodity CPUs, and show that it can yield speedups in end-to-end runtime experiments, both in transfer learning using already-sparsified networks, and in training sparse networks from scratch. Thus, our results provide the first support for sparse training on commodity hardware.

Existing models to extract temporal relations between events lack a principled method to incorporate external knowledge. In this study, we introduce Bayesian-Trans, a Bayesian learning-based method that models the temporal relation representations as latent variables and infers their values via Bayesian inference and translational functions. Compared to conventional neural approaches, instead of performing point estimation to find the best set parameters, the proposed model infers the parameters' posterior distribution directly, enhancing the model's capability to encode and express uncertainty about the predictions. Experimental results on the three widely used datasets show that Bayesian-Trans outperforms existing approaches for event temporal relation extraction. We additionally present detailed analyses on uncertainty quantification, comparison of priors, and ablation studies, illustrating the benefits of the proposed approach.

Specialised pre-trained language models are becoming more frequent in NLP since they can potentially outperform models trained on generic texts. BioBERT (Sanh et al., 2019) and BioClinicalBERT (Alsentzer et al., 2019) are two examples of such models that have shown promise in medical NLP tasks. Many of these models are overparametrised and resource-intensive, but thanks to techniques like Knowledge Distillation (KD), it is possible to create smaller versions that perform almost as well as their larger counterparts. In this work, we specifically focus on development of compact language models for processing clinical texts (i.e. We developed a number of efficient lightweight clinical transformers using knowledge distillation and continual learning, with the number of parameters ranging from 15 million to 65 million. These models performed comparably to larger models such as BioBERT and ClinicalBioBERT and significantly outperformed other compact models trained on general or biomedical data. Our extensive evaluation was done across several standard datasets and covered a wide range of clinical text-mining tasks, including Natural Language Inference, Relation Extraction, Named Entity Recognition, and Sequence Classification. To our knowledge, this is the first comprehensive study specifically focused on creating efficient and compact transformers for clinical NLP tasks. The models and code used in this study can be found on our Huggingface profile at https: //huggingface.co/nlpie and Github page at https://github.com/ Large language models pre-trained on generic texts serve as the foundation upon which most stateof-the-art NLP models are built. There is ample evidence that, for certain domains and downstream tasks, models that are pre-trained on specialised data outperform baselines that have only relied on generic texts (Sanh et al., 2019; Alsentzer et al., 2019; Beltagy et al., 2019; Nguyen et al., 2020; Chalkidis et al., 2020).

Fine-tuning the entire set of parameters of a large pretrained model has become the mainstream approach for transfer learning. To increase its efficiency and prevent catastrophic forgetting and interference, techniques like adapters and sparse fine-tuning have been developed. Adapters are modular, as they can be combined to adapt a model towards different facets of knowledge (e.g., dedicated language and/or task adapters). Sparse fine-tuning is expressive, as it controls the behavior of all model components. In this work, we introduce a new fine-tuning method with both these desirable properties. In particular, we learn sparse, real-valued masks based on a simple variant of the Lottery Ticket Hypothesis. Task-specific masks are obtained from annotated data in a source language, and language-specific masks from masked language modeling in a target language. Both these masks can then be composed with the pretrained model. Unlike adapter-based fine-tuning, this method neither increases the number of parameters at inference time nor alters the original model architecture. Most importantly, it outperforms adapters in zero-shot cross-lingual transfer by a large margin in a series of multilingual benchmarks, including Universal Dependencies, MasakhaNER, and AmericasNLI. Based on an in-depth analysis, we additionally find that sparsity is crucial to prevent both 1) interference between the fine-tunings to be composed and 2) overfitting. We release the code and models at https://github.com/cambridgeltl/composable-sft.

Instrumental variables (IV) have become standard tools for economists, epidemiologists, and social scientists to uncover causal relationships from observational data [3, 46]. Randomization of treatments or policies has been perceived as the gold standard for such tasks, but is generally prohibitive in many real-world scenarios due to time constraints or ethical concerns. When treatment assignment is not randomized, it is generally impossible to discern between the causal effect of treatments and spurious correlations that are induced by unobserved factors. Instead, IVs enable the investigators to incorporate natural variation through an IV that is associated with the treatments, but not with the outcome variable, other than through its effect on the treatments. In economics, for instance, the season-of-birth was used as an IV to study the return from schooling, which measures causal effect of education on labor market earning [16]. In genetic epidemiology, the idea to use genetic variants as IVs, known as Mendelian randomization, has also gained increasing popularity [13, 14].

This paper investigates Graph Neural Networks (GNNs) application for self-supervised network intrusion and anomaly detection. GNNs are a deep learning approach for graph-based data that incorporate graph structures into learning to generalise graph representations and output embeddings. As network flows are naturally graph-based, GNNs are a suitable fit for analysing and learning network behaviour. The majority of current implementations of GNN-based Network Intrusion Detection Systems (NIDSs) rely heavily on labelled network traffic which can not only restrict the amount and structure of input traffic, but also the NIDSs potential to adapt to unseen attacks. To overcome these restrictions, we present Anomal-E, a GNN approach to intrusion and anomaly detection that leverages edge features and graph topological structure in a self-supervised process. This approach is, to the best our knowledge, the first successful and practical approach to network intrusion detection that utilises network flows in a self-supervised, edge leveraging GNN. Experimental results on two modern benchmark NIDS datasets not only clearly display the improvement of using Anomal-E embeddings rather than raw features, but also the potential Anomal-E has for detection on wild network traffic.

Context information is in demand more than ever with the rapid increase in the number of context-aware Internet of Things applications developed worldwide. Research in context and context-awareness is being conducted to broaden its applicability in light of many practical and technical challenges. One of the challenges is improving performance when responding to a large number of context queries. Context Management Platforms that infer and deliver context to applications measure this problem using Quality of Service (QoS) parameters. Although caching is a proven way to improve QoS, transiency of context and features such as variability and heterogeneity of context queries pose an additional real-time cost management problem. This paper presents a critical survey of the state-of-the-art in adaptive data caching with the objective of developing a body of knowledge in cost- and performance-efficient adaptive caching strategies. We comprehensively survey a large number of research publications and evaluate, compare, and contrast different techniques, policies, approaches, and schemes in adaptive caching. Our critical analysis is motivated by the focus on adaptively caching context as a core research problem. A formal definition for adaptive context caching is then proposed, followed by identified features and requirements of a well-designed, objective optimal adaptive context caching strategy.

Performance metrics-driven context caching has a profound impact on throughput and response time in distributed context management systems for real-time context queries. This paper proposes a reinforcement learning based approach to adaptively cache context with the objective of minimizing the cost incurred by context management systems in responding to context queries. Our novel algorithms enable context queries and sub-queries to reuse and repurpose cached context in an efficient manner. This approach is distinctive to traditional data caching approaches by three main features. First, we make selective context cache admissions using no prior knowledge of the context, or the context query load. Secondly, we develop and incorporate innovative heuristic models to calculate expected performance of caching an item when making the decisions. Thirdly, our strategy defines a time-aware continuous cache action space. We present two reinforcement learning agents, a value function estimating actor-critic agent and a policy search agent using deep deterministic policy gradient method. The paper also proposes adaptive policies such as eviction and cache memory scaling to complement our objective. Our method is evaluated using a synthetically generated load of context sub-queries and a synthetic data set inspired from real world data and query samples. We further investigate optimal adaptive caching configurations under different settings. This paper presents, compares, and discusses our findings that the proposed selective caching methods reach short- and long-term cost- and performance-efficiency. The paper demonstrates that the proposed methods outperform other modes of context management such as redirector mode, and database mode, and cache all policy by up to 60% in cost efficiency.

ChatGPT has been a godsend for Joy. The New Zealand-based therapist has ADHD and often struggles with tasks such as drafting difficult emails, with procrastination kicking in when she feels overwhelmed. "Sitting down to compose a complicated email is something I absolutely hate. I would have to use a lot of strategies and accountability to get it done, and I would feel depleted afterward," says Joy, who is in her 30s and lives in Auckland. "But telling GPT'write an email apologising for a delay on an academic manuscript, blame family emergency, ask for consideration for next issue' feels completely doable."