Large language models (LLMs) bear great potential for automating tedious development tasks such as creating and maintaining code documentation. However, it is unclear to what extent developers can effectively prompt LLMs to create concise and useful documentation. We report on a controlled experiment with 20 professionals and 30 computer science students tasked with code documentation generation for two Python functions. The experimental group freely entered ad-hoc prompts in a ChatGPT-like extension of Visual Studio Code, while the control group executed a predefined few-shot prompt. Our results reveal that professionals and students were unaware of or unable to apply prompt engineering techniques. Especially students perceived the documentation produced from ad-hoc prompts as significantly less readable, less concise, and less helpful than documentation from prepared prompts. Some professionals produced higher quality documentation by just including the keyword Docstring in their ad-hoc prompts. While students desired more support in formulating prompts, professionals appreciated the flexibility of ad-hoc prompting. Participants in both groups rarely assessed the output as perfect. Instead, they understood the tools as support to iteratively refine the documentation. Further research is needed to understand which prompting skills and preferences developers have and which support they need for certain tasks.

The unique diagnosis and treatment techniques and remarkable clinical efficacy of traditional Chinese medicine (TCM) make it play an important role in the field of elderly care and healthcare, especially in the rehabilitation of some common chronic diseases of the elderly. Therefore, building a TCM chatbot for healthcare application will help users obtain consultation services in a direct and natural way. However, concepts such as acupuncture points (acupoints) and meridians involved in TCM always appear in the consultation, which cannot be displayed intuitively. To this end, we develop a \textbf{h}ealthcare chat\textbf{bot} (HBot) based on a human body model in 3D and knowledge graph, which provides conversational services such as knowledge Q\&A, prescription recommendation, moxibustion therapy recommendation, and acupoint search. When specific acupoints are involved in the conversations between user and HBot, the 3D body will jump to the corresponding acupoints and highlight them. Moreover, Hbot can also be used in training scenarios to accelerate the teaching process of TCM by intuitively displaying acupuncture points and knowledge cards. The demonstration video is available at https://www.youtube.com/watch?v=UhQhutSKkTU . Our code and dataset are publicly available at Gitee: https://gitee.com/plabrolin/interactive-3d-acup.git

Multi-turn dialogues are a key interaction method between humans and Large Language Models (LLMs), as conversations extend over multiple rounds, keeping LLMs' high generation quality and low latency is a challenge. Mainstream LLMs can be grouped into two categories based on masking strategy: causal LLM and prefix LLM. Several works have demonstrated that prefix LLMs tend to outperform causal ones in scenarios that heavily depend on historical context such as multi-turn dialogues or in-context learning, thanks to their bidirectional attention on prefix sequences. However, prefix LLMs have an inherent inefficient training problem in multi-turn dialogue datasets. In addition, the attention mechanism of prefix LLM makes it unable to reuse Key-Value Cache (KV Cache) across dialogue rounds to reduce generation latency. In this paper, we propose a novel masking scheme called Intermittent Semi-working Mask (ISM) to address these problems. Specifically, we apply alternate bidirectional and unidirectional attention on queries and answers in the dialogue history. In this way, ISM is able to maintain the high quality of prefix LLM and low generation latency of causal LLM, simultaneously. Extensive experiments illustrate that our ISM achieves significant performance.

Large-scale text-to-speech (TTS) models have made significant progress recently.However, they still fall short in the generation of Chinese dialectal speech. Toaddress this, we propose Bailing-TTS, a family of large-scale TTS models capable of generating high-quality Chinese dialectal speech. Bailing-TTS serves as a foundation model for Chinese dialectal speech generation. First, continual semi-supervised learning is proposed to facilitate the alignment of text tokens and speech tokens. Second, the Chinese dialectal representation learning is developed using a specific transformer architecture and multi-stage training processes. With the proposed design of novel network architecture and corresponding strategy, Bailing-TTS is able to generate Chinese dialectal speech from text effectively and efficiently. Experiments demonstrate that Bailing-TTS generates Chinese dialectal speech towards human-like spontaneous representation. Readers are encouraged to listen to demos at \url{https://c9412600.github.io/bltts_tech_report/index.html}.

Some companies(e.g., Microsoft Research and Google DeepMind) have discovered some of the limitations of GPTs autoregressive paradigm next-word prediction, manifested in the model lack of planning, working memory, backtracking, and reasoning skills. GPTs rely on a local and greedy process of generating the next word, without a global understanding of the task or the output.We have confirmed the above limitations through specialized empirical studies of code comprehension. Although GPT4 is good at producing fluent and coherent text, it cannot handle complex logic and generate new code that haven not been seen, and it relies too much on the formatting of the prompt to generate the correct code.We propose a new paradigm for code understanding that goes beyond the next-word prediction paradigm, inspired by the successful application of diffusion techniques to image generation(Dalle2, Sora) and protein structure generation(AlphaFold3), which have no autoregressive constraints.Instead of encoding the code in a form that mimics natural language, we encode the code as a heterogeneous image paradigm with a memory of global information that mimics both images and protein structures.We then refer to Sora's CLIP upstream text-to-image encoder model to design a text-to-code encoder model that can be applied to various downstream code understanding tasks.The model learns the global understanding of code under the new paradigm heterogeneous image, connects the encoding space of text and code, and encodes the input of text into the vector of code most similar to it.Using self-supervised comparative learning on 456,360 text-code pairs, the model achieved a zero-shot prediction of new data. This work is the basis for future work on code generation using diffusion techniques under a new paradigm to avoid autoregressive limitations.

The ability of generative large language models (LLMs) to perform in-context learning has given rise to a large body of research into how best to prompt models for various natural language processing tasks. In this paper, we focus on machine translation (MT), a task that has been shown to benefit from in-context translation examples. However no systematic studies have been published on how best to select examples, and mixed results have been reported on the usefulness of similarity-based selection over random selection. We provide a study covering multiple LLMs and multiple in-context example retrieval strategies, comparing multilingual sentence embeddings. We cover several language directions, representing different levels of language resourcedness (English into French, German, Swahili and Wolof). Contrarily to previously published results, we find that sentence embedding similarity can improve MT, especially for low-resource language directions, and discuss the balance between selection pool diversity and quality. We also highlight potential problems with the evaluation of LLM-based MT and suggest a more appropriate evaluation protocol, adapting the COMET metric to the evaluation of LLMs. Code and outputs are freely available at https://github.com/ArmelRandy/ICL-MT.

The first study of its kind has determined what many have long suspected - AI left-wing. A total of 24 Large Language Models (LLMs), including Google's Gemini, OpenAI's ChatGPT and even Elon Musk's Grok, were asked political charged questions during tests of its values, party affiliation and personality. The results showed the all LLMs produced answers that were largely'Progressive,' 'Democratic' and'Green,' and included values like'Equality,' 'World' and'Progress.' The researcher raised concern about companies integrating AI into products like search engines such as Google that has come under fire its Chrome that Donald Trump and Elon Musk claimed is interfering with the election. The results showed the all LLMs produced answers that were largely'Progressive,' 'Democratic' and'Green,' and included values like'Equality,' 'World' and'Progress' Chrome uses AI to auto-complete results, but last week it was found when users typed in assassination attempt on,' the browser suggested former President Ronald Reagan, Bob Marley, and other figures.

Realizing AI's full potential on a mass scale will require more than just executives' enthusiasm; becoming a truly AI-first enterprise will require a significant, sustained investment in cloud infrastructure and strategy. In 2024, the cloud has evolved beyond its initial purpose as a storage tool and cost saver to become a crucial driver of innovation, transformation, and disruption. Now, with AI in the mix, enterprises are looking to the cloud to support large language models (LLMs) to maximize R&D performance and prevent cybersecurity attacks, among other high-impact use cases. A 2023 report by Infosys looks at how prepared companies are to realize the combined potential of cloud and AI. To further assess this state of readiness, MIT Technology Review Insights and Infosys surveyed 500 business leaders across industries such as IT, manufacturing, financial services, and consumer goods about how their organizations are thinking about--and acting upon--an integrated cloud and AI strategy.

The EU Artificial Intelligence Act (AI Act) came into force in the European Union (EU) on 1 August 2024. It is a key piece of legislation both for the citizens at the heart of AI technologies and for the industry active in the internal market. The AI Act imposes progressive compliance on organisations - both private and public - involved in the global value chain of AI systems and models marketed and used in the EU. While the Act is unprecedented on an international scale in terms of its horizontal and binding regulatory scope, its global appeal in support of trustworthy AI is one of its major challenges.

Foundation models have had a big impact in recent years and billions of dollars are being invested in them in the current AI boom. The more popular ones, such as Chat-GPT, are trained on large amounts of data from the Internet, and then reinforcement learning, RAG, prompt engineering and cognitive modelling are used to fine-tune and augment their behavior. This technology has been used to create models of individual people, such as Caryn Marjorie. However, these chatbots are not based on people's actual emotional and physiological responses to their environment, so they are, at best, surface-level approximations to the characters they are imitating. This paper describes how a new type of foundation model - a first-person foundation model - could be created from recordings of what a person sees and hears as well as their emotional and physiological reactions to these stimuli. A first-person foundation model would map environmental stimuli to a person's emotional and physiological states, and map a person's emotional and physiological states to their behavior. First-person foundation models have many exciting applications, including a new type of recommendation engine, personal assistants, generative adversarial networks, dating and recruitment. To obtain training data for a first-person foundation model, we have developed a recording rig that captures what the wearer is seeing and hearing as well as their emotional and physiological states. This novel source of data could help to address the shortage of new data for building the next generation of foundation models.