We provide a birds eye view of the rapid developments in AI and Deep Learning that has led to the path-breaking emergence of AI in Large Language Models. The aim of this study is to place all these developments in a pragmatic broader historical social perspective without any exaggerations while at the same time without any pessimism that created the AI winter in the 1970s to 1990s. We also at the same time point out toxicity, bias, memorization, sycophancy, logical inconsistencies, hallucinations that exist just as a warning to the overly optimistic. We note here that just as this emergence of AI seems to occur at a threshold point in the number of neural connections or weights, it has also been observed that human brain and especially the cortex region is nothing special or extraordinary but simply a case of scaled-up version of the primate brain and that even the human intelligence seems like an emergent phenomena of scale.
The speech-to-song illusion is a robust psychological phenomenon whereby a spoken sentence sounds increasingly more musical as it is repeated. Despite decades of research, a complete formal account of this transformation is still lacking, and some of its nuanced characteristics, namely, that certain phrases appear to transform while others do not, is not well understood. Here we provide a formal account of this phenomenon, by recasting it as a statistical inference whereby a rational agent attempts to decide whether a sequence of utterances is more likely to have been produced in a song or speech. Using this approach and analyzing song and speech corpora, we further introduce a novel prose-to-lyrics illusion that is purely text-based. In this illusion, simply duplicating written sentences makes them appear more like song lyrics. We provide robust evidence for this new illusion in both human participants and large language models.
Vision-language models (VLMs) have emerged as formidable tools, showing their strong capability in handling various open-vocabulary tasks in image recognition, text-driven visual content generation, and visual chatbots, to name a few. In recent years, considerable efforts and resources have been devoted to adaptation methods for improving downstream performance of VLMs, particularly on parameter-efficient fine-tuning methods like prompt learning. However, a crucial aspect that has been largely overlooked is the confidence calibration problem in fine-tuned VLMs, which could greatly reduce reliability when deploying such models in the real world. This paper bridges the gap by systematically investigating the confidence calibration problem in the context of prompt learning and reveals that existing calibration methods are insufficient to address the problem, especially in the open-vocabulary setting. To solve the problem, we present a simple and effective approach called Distance-Aware Calibration (DAC), which is based on scaling the temperature using as guidance the distance between predicted text labels and base classes. The experiments with 7 distinct prompt learning methods applied across 11 diverse downstream datasets demonstrate the effectiveness of DAC, which achieves high efficacy without sacrificing the inference speed.
The robustness of large language models (LLMs) becomes increasingly important as their use rapidly grows in a wide range of domains. Retrieval-Augmented Generation (RAG) is considered as a means to improve the trustworthiness of text generation from LLMs. However, how the outputs from RAG-based LLMs are affected by slightly different inputs is not well studied. In this work, we find that the insertion of even a short prefix to the prompt leads to the generation of outputs far away from factually correct answers. We systematically evaluate the effect of such prefixes on RAG by introducing a novel optimization technique called Gradient Guided Prompt Perturbation (GGPP). GGPP achieves a high success rate in steering outputs of RAG-based LLMs to targeted wrong answers. It can also cope with instructions in the prompts requesting to ignore irrelevant context. We also exploit LLMs' neuron activation difference between prompts with and without GGPP perturbations to give a method that improves the robustness of RAG-based LLMs through a highly effective detector trained on neuron activation triggered by GGPP generated prompts. Our evaluation on open-sourced LLMs demonstrates the effectiveness of our methods.
Reasoning is a fundamental component for achieving language understanding. Among the multiple types of reasoning, conditional reasoning, the ability to draw different conclusions depending on some condition, has been understudied in large language models (LLMs). Recent prompting methods, such as chain of thought, have significantly improved LLMs on reasoning tasks. Nevertheless, there is still little understanding of what triggers reasoning abilities in LLMs. We hypothesize that code prompts can trigger conditional reasoning in LLMs trained on text and code. We propose a chain of prompts that transforms a natural language problem into code and prompts the LLM with the generated code. Our experiments find that code prompts exhibit a performance boost between 2.6 and 7.7 points on GPT 3.5 across multiple datasets requiring conditional reasoning. We then conduct experiments to discover how code prompts elicit conditional reasoning abilities and through which features. We observe that prompts need to contain natural language text accompanied by high-quality code that closely represents the semantics of the instance text. Furthermore, we show that code prompts are more efficient, requiring fewer demonstrations, and that they trigger superior state tracking of variables or key entities.
Large Language Models (LLMs) are one of the most promising technologies for the next era of speech generation systems, due to their scalability and in-context learning capabilities. Nevertheless, they suffer from multiple stability issues at inference time, such as hallucinations, content skipping or speech repetitions. In this work, we introduce a new self-supervised Voice Conversion (VC) architecture which can be used to learn to encode transitory features, such as content, separately from stationary ones, such as speaker ID or recording conditions, creating speaker-disentangled representations. Using speaker-disentangled codes to train LLMs for text-to-speech (TTS) allows the LLM to generate the content and the style of the speech only from the text, similarly to humans, while the speaker identity is provided by the decoder of the VC model. Results show that LLMs trained over speaker-disentangled self-supervised representations provide an improvement of 4.7pp in speaker similarity over SOTA entangled representations, and a word error rate (WER) 5.4pp lower. Furthermore, they achieve higher naturalness than human recordings of the LibriTTS test-other dataset. Finally, we show that using explicit reference embedding negatively impacts intelligibility (stability), with WER increasing by 14pp compared to the model that only uses text to infer the style.
In the pursuit of developing Large Language Models (LLMs) that adhere to societal standards, it is imperative to discern the existence of toxicity in the generated text. The majority of existing toxicity metrics rely on encoder models trained on specific toxicity datasets. However, these encoders are susceptible to out-of-distribution (OOD) problems and depend on the definition of toxicity assumed in a dataset. In this paper, we introduce an automatic robust metric grounded on LLMs to distinguish whether model responses are toxic. We start by analyzing the toxicity factors, followed by examining the intrinsic toxic attributes of LLMs to ascertain their suitability as evaluators. Subsequently, we evaluate our metric, LLMs As ToxiciTy Evaluators (LATTE), on evaluation datasets.The empirical results indicate outstanding performance in measuring toxicity, improving upon state-of-the-art metrics by 12 points in F1 score without training procedure. We also show that upstream toxicity has an influence on downstream metrics.
Large Language Model (LLM) assistants, such as ChatGPT, have emerged as potential alternatives to search methods for helping users navigate complex, feature-rich software. LLMs use vast training data from domain-specific texts, software manuals, and code repositories to mimic human-like interactions, offering tailored assistance, including step-by-step instructions. In this work, we investigated LLM-generated software guidance through a within-subject experiment with 16 participants and follow-up interviews. We compared a baseline LLM assistant with an LLM optimized for particular software contexts, SoftAIBot, which also offered guidelines for constructing appropriate prompts. We assessed task completion, perceived accuracy, relevance, and trust. Surprisingly, although SoftAIBot outperformed the baseline LLM, our results revealed no significant difference in LLM usage and user perceptions with or without prompt guidelines and the integration of domain context. Most users struggled to understand how the prompt's text related to the LLM's responses and often followed the LLM's suggestions verbatim, even if they were incorrect. This resulted in difficulties when using the LLM's advice for software tasks, leading to low task completion rates. Our detailed analysis also revealed that users remained unaware of inaccuracies in the LLM's responses, indicating a gap between their lack of software expertise and their ability to evaluate the LLM's assistance. With the growing push for designing domain-specific LLM assistants, we emphasize the importance of incorporating explainable, context-aware cues into LLMs to help users understand prompt-based interactions, identify biases, and maximize the utility of LLM assistants.
In this paper, we present a new embedding model, called M3-Embedding, which is distinguished for its versatility in Multi-Linguality, Multi-Functionality, and Multi-Granularity. It can support more than 100 working languages, leading to new state-of-the-art performances on multi-lingual and cross-lingual retrieval tasks. It can simultaneously perform the three common retrieval functionalities of embedding model: dense retrieval, multi-vector retrieval, and sparse retrieval, which provides a unified model foundation for real-world IR applications. It is able to process inputs of different granularities, spanning from short sentences to long documents of up to 8192 tokens. The effective training of M3-Embedding involves the following technical contributions. We propose a novel self-knowledge distillation approach, where the relevance scores from different retrieval functionalities can be integrated as the teacher signal to enhance the training quality. We also optimize the batching strategy, enabling a large batch size and high training throughput to ensure the discriminativeness of embeddings. To the best of our knowledge, M3-Embedding is the first embedding model which realizes such a strong versatility. The model and code will be publicly available at https://github.com/FlagOpen/FlagEmbedding.
Matching patients to clinical trials is a key unsolved challenge in bringing new drugs to market. Today, identifying patients who meet a trial's eligibility criteria is highly manual, taking up to 1 hour per patient. Automated screening is challenging, however, as it requires understanding unstructured clinical text. Large language models (LLMs) offer a promising solution. In this work, we explore their application to trial matching. First, we design an LLM-based system which, given a patient's medical history as unstructured clinical text, evaluates whether that patient meets a set of inclusion criteria (also specified as free text). Our zero-shot system achieves state-of-the-art scores on the n2c2 2018 cohort selection benchmark. Second, we improve the data and cost efficiency of our method by identifying a prompting strategy which matches patients an order of magnitude faster and more cheaply than the status quo, and develop a two-stage retrieval pipeline that reduces the number of tokens processed by up to a third while retaining high performance. Third, we evaluate the interpretability of our system by having clinicians evaluate the natural language justifications generated by the LLM for each eligibility decision, and show that it can output coherent explanations for 97% of its correct decisions and 75% of its incorrect ones. Our results establish the feasibility of using LLMs to accelerate clinical trial operations.