Moderate-sized large language models (LLMs) -- those with 7B or 13B parameters -- exhibit promising machine translation (MT) performance. However, even the top-performing 13B LLM-based translation models, like ALMA, does not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4. In this study, we bridge this performance gap. We first assess the shortcomings of supervised fine-tuning for LLMs in the MT task, emphasizing the quality issues present in the reference data, despite being human-generated. Then, in contrast to SFT which mimics reference translations, we introduce Contrastive Preference Optimization (CPO), a novel approach that trains models to avoid generating adequate but not perfect translations. Applying CPO to ALMA models with only 22K parallel sentences and 12M parameters yields significant improvements. The resulting model, called ALMA-R, can match or exceed the performance of the WMT competition winners and GPT-4 on WMT'21, WMT'22 and WMT'23 test datasets.

Mental health challenges pose considerable global burdens on individuals and communities. Recent data indicates that more than 20% of adults may encounter at least one mental disorder in their lifetime. On the one hand, the advancements in large language models have facilitated diverse applications, yet a significant research gap persists in understanding and enhancing the potential of large language models within the domain of mental health. On the other hand, across various applications, an outstanding question involves the capacity of large language models to comprehend expressions of human mental health conditions in natural language. This study presents an initial evaluation of large language models in addressing this gap. Due to this, we compare the performance of Llama-2 and ChatGPT with classical Machine as well as Deep learning models. Our results on the DAIC-WOZ dataset show that transformer-based models, like BERT or XLNet, outperform the large language models.

The utilization of long contexts poses a big challenge for LLMs due to their limited context window size. Although the context window can be extended through fine-tuning, it will result in a considerable cost at both training and inference time, and exert an unfavorable impact to the LLM's original capabilities. In this work, we propose a new method called Activation Beacon, which condenses LLM's raw activations into compact forms such that the LLM can perceive a longer context with a limited context window. Activation Beacon is introduced as a plug-in module, which fully preserves the LLM's original capability in short contexts. It works with the sliding window to streamingly process the long context, which leads to a competitive memory and time efficiency in both training and inference. Activation Beacon is trained with short-sequence data of diversified condensing ratios. Thanks to such a treatment, it can be effectively learned to support different context lengths with a small training cost. Our experiment verifies Activation Beacon's effectiveness of context extension: it can remarkably accomplish high-quality extension of Llama-2-7B's context by $\times100$ times (from 4K to 400K); meanwhile, it can also achieve superior performances across a variety of long-context language modeling and understanding tasks. The source code and model checkpoint are available at \url{https://github.com/FlagOpen/FlagEmbedding}.

Loss spikes often occur during pre-training of large language models. The spikes degrade the performance of large language models and sometimes ruin the pre-training. Since the pre-training needs a vast computational budget, we should avoid such spikes. To investigate the cause of loss spikes, we focus on gradients of internal layers. Through theoretical analyses, we reveal two causes of the exploding gradients, and provide requirements to prevent the explosion. In addition, we propose a method to satisfy the requirements by combining the initialization method and a simple modification to embeddings. We conduct various experiments to verify our theoretical analyses empirically. Experimental results indicate that the combination is effective in preventing spikes during pre-training.

Post-training quantization (PTQ) has played a key role in compressing large language models (LLMs) with ultra-low costs. However, existing PTQ methods only focus on handling the outliers within one layer or one block, which ignores the dependency of blocks and leads to severe performance degradation in low-bit settings. In this paper, we propose CBQ, a cross-block reconstruction-based PTQ method for LLMs. CBQ employs a cross-block dependency using a homologous reconstruction scheme, establishing long-range dependencies across multiple blocks to minimize error accumulation. Furthermore, CBQ incorporates a coarse-to-fine preprocessing (CFP) strategy for suppressing weight and activation outliers, coupled with an adaptive LoRA-Rounding technique for precise weight quantization. These innovations enable CBQ to not only handle extreme outliers effectively but also improve overall quantization accuracy. Extensive experiments show that CBQ achieves superior low-bit quantization (W4A4, W4A8, W2A16) and outperforms existing state-of-the-art methods across various LLMs and datasets. Notably, CBQ quantizes the 4-bit LLAMA1-65B model within only 4.3 hours on a single GPU, achieving a commendable tradeoff between performance and quantization efficiency.

Rapid advancements in large language models (LLMs) have revitalized in LLM-based agents, exhibiting impressive human-like behaviors and cooperative capabilities in various scenarios. However, these agents also bring some exclusive risks, stemming from the complexity of interaction environments and the usability of tools. This paper delves into the safety of LLM-based agents from three perspectives: agent quantity, role definition, and attack level. Specifically, we initially propose to employ a template-based attack strategy on LLM-based agents to find the influence of agent quantity. In addition, to address interaction environment and role specificity issues, we introduce Evil Geniuses (EG), an effective attack method that autonomously generates prompts related to the original role to examine the impact across various role definitions and attack levels. EG leverages Red-Blue exercises, significantly improving the generated prompt aggressiveness and similarity to original roles. Our evaluations on CAMEL, Metagpt and ChatDev based on GPT-3.5 and GPT-4, demonstrate high success rates. Extensive evaluation and discussion reveal that these agents are less robust, prone to more harmful behaviors, and capable of generating stealthier content than LLMs, highlighting significant safety challenges and guiding future research. Our code is available at https://github.com/T1aNS1R/Evil-Geniuses.

Large language models (LLMs), such as ChatGPT, have received substantial attention due to their capabilities for understanding and generating human language. LLMs in medicine to assist physicians for patient care are emerging as a promising research direction in both artificial intelligence and clinical medicine. This review provides a comprehensive overview of the principles, applications, and challenges faced by LLMs in medicine. We address the following specific questions: 1) How should medical LLMs be built? 2) What are the measures for the downstream performance of medical LLMs? 3) How should medical LLMs be utilized in real-world clinical practice? 4) What challenges arise from the use of medical LLMs? and 5) How should we better construct and utilize medical LLMs? This review aims to provide insights into the opportunities and challenges of LLMs in medicine, and serve as a practical resource for constructing effective medical LLMs. We also maintain and regularly updated list of practical guides on medical LLMs at https://github.com/AI-in-Health/MedLLMsPracticalGuide.

Sampling biases can cause distribution shifts between train and test datasets for supervised learning tasks, obscuring our ability to understand the generalization capacity of a model. This is especially important considering the wide adoption of pre-trained foundational neural networks -- whose behavior remains poorly understood -- for transfer learning (TL) tasks. We present a case study for TL on the Sentiment140 dataset and show that many pre-trained foundation models encode different representations of Sentiment140's manually curated test set M from the automatically labeled training set P, confirming that a distribution shift has occurred. We argue training on P and measuring performance on M is a biased measure of generalization. Experiments on pre-trained GPT-2 show that the features learnable from P do not improve (and in fact hamper) performance on M. Linear probes on pre-trained GPT-2's representations are robust and may even outperform overall fine-tuning, implying a fundamental importance for discerning distribution shift in train/test splits for model interpretation. Foundation models Brown et al. (2020); Touvron et al. (2023); Liu et al. (2023); Rombach et al. (2021) have quickly integrated themselves into the standard machine learning development stack, in particular for their adaptability to specialized tasks Zhai et al. (2023); Wang et al. (2020).

Most NLP tasks are modeled as supervised learning and thus require labeled training data to train effective models. However, manually producing such data at sufficient quality and quantity is known to be costly and time-intensive. Current research addresses this bottleneck by exploring a novel paradigm called zero-shot learning via dataset generation. Here, a powerful LLM is prompted with a task description to generate labeled data that can be used to train a downstream NLP model. For instance, an LLM might be prompted to "generate 500 movie reviews with positive overall sentiment, and another 500 with negative sentiment." The generated data could then be used to train a binary sentiment classifier, effectively leveraging an LLM as a teacher to a smaller student model. With this demo, we introduce Fabricator, an open-source Python toolkit for dataset generation. Fabricator implements common dataset generation workflows, supports a wide range of downstream NLP tasks (such as text classification, question answering, and entity recognition), and is integrated with well-known libraries to facilitate quick experimentation. With Fabricator, we aim to support researchers in conducting reproducible dataset generation experiments using LLMs and help practitioners apply this approach to train models for downstream tasks.

Never having seen an object and heard its sound simultaneously, can the model still accurately localize its visual position from the input audio? In this work, we concentrate on the Audio-Visual Localization and Segmentation tasks but under the demanding zero-shot and few-shot scenarios. To achieve this goal, different from existing approaches that mostly employ the encoder-fusion-decoder paradigm to decode localization information from the fused audio-visual feature, we introduce the encoder-prompt-decoder paradigm, aiming to better fit the data scarcity and varying data distribution dilemmas with the help of abundant knowledge from pre-trained models. Specifically, we first propose to construct Semantic-aware Audio Prompt (SAP) to help the visual foundation model focus on sounding objects, meanwhile, the semantic gap between the visual and audio modalities is also encouraged to shrink. Then, we develop a Correlation Adapter (ColA) to keep minimal training efforts as well as maintain adequate knowledge of the visual foundation model. By equipping with these means, extensive experiments demonstrate that this new paradigm outperforms other fusion-based methods in both the unseen class and cross-dataset settings. We hope that our work can further promote the generalization study of Audio-Visual Localization and Segmentation in practical application scenarios.