Alignment is crucial for training large language models. The predominant strategy is Reinforcement Learning from Human Feedback (RLHF), with Proximal Policy Optimization (PPO) as the de-facto algorithm. Yet, PPO is known to struggle with computational inefficiency, a challenge that this paper aims to address. We identify three important properties of RLHF tasks: fast simulation, deterministic transitions, and trajectory-level rewards, which are not leveraged in PPO. Based on these properties, we develop ReMax, a new algorithm tailored for RLHF. The design of ReMax builds on the celebrated algorithm REINFORCE but is enhanced with a new variance-reduction technique. ReMax offers threefold advantages over PPO: first, it is simple to implement with just 6 lines of code. It further eliminates more than 4 hyper-parameters in PPO, which are laborious to tune. Second, ReMax reduces memory usage by about 50%. To illustrate, PPO runs out of memory when fine-tuning a Llama2-7B model on A100-80GB GPUs, whereas ReMax can support the training. Even though memory-efficient techniques (e.g., ZeRO and offload) are employed for PPO to afford training, ReMax can utilize a larger batch size to increase throughput. Third, in terms of wall-clock time, PPO is about twice as slow as ReMax per iteration. Importantly, these improvements do not sacrifice task performance. We hypothesize that these advantages can be maintained in larger-scale models.

This survey addresses the crucial issue of factuality in Large Language Models (LLMs). As LLMs find applications across diverse domains, the reliability and accuracy of their outputs become vital. We define the Factuality Issue as the probability of LLMs to produce content inconsistent with established facts. We first delve into the implications of these inaccuracies, highlighting the potential consequences and challenges posed by factual errors in LLM outputs. Subsequently, we analyze the mechanisms through which LLMs store and process facts, seeking the primary causes of factual errors. Our discussion then transitions to methodologies for evaluating LLM factuality, emphasizing key metrics, benchmarks, and studies. We further explore strategies for enhancing LLM factuality, including approaches tailored for specific domains. We focus two primary LLM configurations standalone LLMs and Retrieval-Augmented LLMs that utilizes external data, we detail their unique challenges and potential enhancements. Our survey offers a structured guide for researchers aiming to fortify the factual reliability of LLMs.

We introduce a new zero resource code-switched speech benchmark designed to directly assess the code-switching capabilities of self-supervised speech encoders. We showcase a baseline system of language modeling on discrete units to demonstrate how the code-switching abilities of speech encoders can be assessed in a zero-resource manner. Our experiments encompass a variety of well-known speech encoders, including Wav2vec 2.0, HuBERT, XLSR, etc. We examine the impact of pre-training languages and model size on benchmark performance. Notably, though our results demonstrate that speech encoders with multilingual pre-training, exemplified by XLSR, outperform monolingual variants (Wav2vec 2.0, HuBERT) in code-switching scenarios, there is still substantial room for improvement in their code-switching linguistic abilities.

Currently, dialogue systems have achieved high performance in processing text-based communication. However, they have not yet effectively incorporated visual information, which poses a significant challenge. Furthermore, existing models that incorporate images in dialogue generation focus on discussing the image itself. Our proposed approach presents a novel perspective on multi-modal dialogue systems, which interprets the image in the context of the dialogue. By doing so, we aim to expand the capabilities of current dialogue systems and transition them from single modality (text) to multi-modality. However, there is a lack of validated English datasets that contain both images and dialogue contexts for this task. Thus, we propose a two-stage approach to automatically construct a multi-modal dialogue dataset. In the first stage, we utilize text-to-image similarity and sentence similarity to identify which utterances could be replaced with an image. In the second stage, we replace those utterances by selecting a subset of relevant images and filtering them with a visual question answering model. We used this approach, along with additional labeling, to create the IMage Augmented multi-modal Dialogue dataset (IMAD), which can serve as a validated dataset for this task. Furthermore, we propose a baseline model trained on this dataset, which outperforms model trained on the same data without images and BlenderBot.

During the preceding biennium, vision-language pre-training has achieved noteworthy success on several downstream tasks. Nevertheless, acquiring high-quality image-text pairs, where the pairs are entirely exclusive of each other, remains a challenging task, and noise exists in the commonly used datasets. To address this issue, we propose SoftCLIP, a novel approach that relaxes the strict one-to-one constraint and achieves a soft cross-modal alignment by introducing a softened target, which is generated from the fine-grained intra-modal self-similarity. The intra-modal guidance is indicative to enable two pairs have some local similarities and model many-to-many relationships between the two modalities. Besides, since the positive still dominates in the softened target distribution, we disentangle the negatives in the distribution to further boost the relation alignment with the negatives in the cross-modal learning. Extensive experiments demonstrate the effectiveness of SoftCLIP. In particular, on ImageNet zero-shot classification task, using CC3M/CC12M as pre-training dataset, SoftCLIP brings a top-1 accuracy improvement of 6.8%/7.2% over the CLIP baseline.

Deformable object manipulation (DOM) for robots has a wide range of applications in various fields such as industrial, service and health care sectors. However, compared to manipulation of rigid objects, DOM poses significant challenges for robotic perception, modeling and manipulation, due to the infinite dimensionality of the state space of deformable objects (DOs) and the complexity of their dynamics. The development of computer graphics and machine learning has enabled novel techniques for DOM. These techniques, based on data-driven paradigms, can address some of the challenges that analytical approaches of DOM face. However, some existing reviews do not include all aspects of DOM, and some previous reviews do not summarize data-driven approaches adequately. In this article, we survey more than 150 relevant studies (data-driven approaches mainly) and summarize recent advances, open challenges, and new frontiers for aspects of perception, modeling and manipulation for DOs. Particularly, we summarize initial progress made by Large Language Models (LLMs) in robotic manipulation, and indicates some valuable directions for further research. We believe that integrating data-driven approaches and analytical approaches can provide viable solutions to open challenges of DOM.

Large Language Models (LLMs) are popular for their impressive abilities, but the need for model-specific fine-tuning or task-specific prompt engineering can hinder their generalization. We propose UPRISE (Universal Prompt Retrieval for Improving zero-Shot Evaluation), which tunes a lightweight and versatile retriever that automatically retrieves prompts for a given zero-shot task input. Specifically, we demonstrate universality in a cross-task and cross-model scenario: the retriever is tuned on a diverse set of tasks, but tested on unseen task types; we use a small frozen LLM, GPT-Neo-2.7B, for tuning the retriever, but test the retriever on different LLMs of much larger scales, such as BLOOM-7.1B, OPT-66B and GPT3-175B. Additionally, we show that UPRISE mitigates the hallucination problem in our experiments with ChatGPT, suggesting its potential to improve even the strongest LLMs. Our model and code are available at https://github.com/microsoft/LMOps.

In November 2022, OpenAI launched ChatGPT. Within five days, it had over a million users. Six months later, the CEOs of the world's leading AI companies, and hundreds of researchers and experts, signed a short statement warning that mitigating the risk of extinction from AI should be a global priority on the scale of preventing nuclear war. AI's rapid technological progress and the dire warnings from its creators provoked a reaction in capitals around the world. But as lawmakers and regulators rushed to write the rules charting AI's future, many warned their efforts were insufficient to mitigate the risks from, and capitalize on the benefits of AI.

AI resume-screening systems used by major Fortune 500 companies may be discriminating against mothers, a study suggests. Researchers at New York University believe they may have unearthed a bias against women who have taken considerable time off work for maternity leave. The team fed hundreds of resumes to four models, including ChatGPT and Google's Bard, and found that they all rejected resumes with the gap. When asked for the reasoning behind the decision, the tech shared: 'Including personal information about maternity leave is not relevant to the job and could be seen as a liability.' The researchers described the trends as'alarming,' given that virtually every major company uses the tech to screen resumes.

The news: Google DeepMind has used a large language model to crack a famous unsolved problem in pure mathematics. The researchers say it is the first time a large language model has been used to discover a solution to a long-standing scientific puzzle--producing verifiable and valuable new information that did not previously exist. Why it matters: Large language models have a reputation for making things up, not for providing new facts. Google DeepMind's new tool, called FunSearch, could change that. It shows that they can indeed make discoveries--if they are coaxed just so, and if you throw out the majority of what they come up with.