A ChatGPT-like system for drug compounds could be a game-changer in pharmaceutical research, accelerating drug discovery, enhancing our understanding of structure-activity relationships, guiding lead optimization, aiding drug repurposing, reducing the failure rate, and streamlining clinical trials. In this work, we make an initial attempt towards enabling ChatGPT-like capabilities on drug molecule graphs, by developing a prototype system DrugChat. DrugChat works in a similar way as ChatGPT. Users upload a compound molecule graph and ask various questions about this compound. DrugChat will answer these questions in a multi-turn, interactive manner. The DrugChat system consists of a graph neural network (GNN), a large language model (LLM), and an adaptor. The GNN takes a compound molecule graph as input and learns a representation for this graph. The adaptor transforms the graph representation produced by the GNN into another representation that is acceptable to the LLM. The LLM takes the compound representation transformed by the adaptor and users' questions about this compound as inputs and generates answers. All these components are trained end-to-end. To train DrugChat, we collected instruction tuning datasets which contain 10,834 drug compounds and 143,517 question-answer pairs. The code and data is available at \url{https://github.com/UCSD-AI4H/drugchat}

Abstract-- Large language models have been useful in expanding mental health care delivery. ChatGPT, in particular, has gained popularity for its ability to generate human-like dialogue. To enable its utilization, we propose a text ambiguation framework that preserves user privacy. We ground this in the task of addressing stress prompted by user-provided texts to demonstrate the viability and helpfulness of privacy-preserved generations and find that recommendations are able to be moderately helpful and relevant, even if original user text is not used. We measured cosine similarity after calculating TF-IDF on Language technologies have proven useful in improving P versus NP responses and found an average score of 0.25, mental health outcomes according to scholarly literature [1], indicating some similarity between responses.

The primary way of building AI applications is shifting from training specialist models to prompting generalist models. A common practice for prompting generalist models, often referred to as in-context learning, is to append a few examples (demonstrations) to the prompt to help the model better understand the task. While effective, in-context learning can be inefficient because it makes the input prompt much longer, consuming valuable space in the context window and leading to larger computational costs. In this paper, we propose DynaICL, a recipe for efficient prompting with black-box generalist models that dynamically allocate in-context examples according to the input complexity and the computational budget. To achieve this, we train a meta controller that predicts the number of in-context examples suitable for the generalist model to make a good prediction based on the performance-efficiency trade-off for a specific input. We then dynamically allocate the number of demonstrations for an input according to predictions from the meta controller and the given computation budget. Experimental results show that dynamic example allocation helps achieve a better performance-efficiency trade-off in two practical settings where computational resources or the required performance is constrained. Specifically, DynaICL saves up to 46% token budget compared to the common practice that allocates the same number of in-context examples to each input. We also find that a meta controller trained on a certain backbone model and tasks can successfully generalize to unseen models and tasks.

Objective: We aim to develop an open-source natural language processing (NLP) package, SODA (i.e., SOcial DeterminAnts), with pre-trained transformer models to extract social determinants of health (SDoH) for cancer patients, examine the generalizability of SODA to a new disease domain (i.e., opioid use), and evaluate the extraction rate of SDoH using cancer populations. Methods: We identified SDoH categories and attributes and developed an SDoH corpus using clinical notes from a general cancer cohort. We compared four transformer-based NLP models to extract SDoH, examined the generalizability of NLP models to a cohort of patients prescribed with opioids, and explored customization strategies to improve performance. We applied the best NLP model to extract 19 categories of SDoH from the breast (n=7,971), lung (n=11,804), and colorectal cancer (n=6,240) cohorts. Results and Conclusion: We developed a corpus of 629 cancer patients notes with annotations of 13,193 SDoH concepts/attributes from 19 categories of SDoH. The Bidirectional Encoder Representations from Transformers (BERT) model achieved the best strict/lenient F1 scores of 0.9216 and 0.9441 for SDoH concept extraction, 0.9617 and 0.9626 for linking attributes to SDoH concepts. Fine-tuning the NLP models using new annotations from opioid use patients improved the strict/lenient F1 scores from 0.8172/0.8502 to 0.8312/0.8679. The extraction rates among 19 categories of SDoH varied greatly, where 10 SDoH could be extracted from >70% of cancer patients, but 9 SDoH had a low extraction rate (<70% of cancer patients). The SODA package with pre-trained transformer models is publicly available at https://github.com/uf-hobiinformatics-lab/SDoH_SODA.

Fox News Flash top sports headlines are here. Check out what's clicking on Foxnews.com. Pro Football Hall of Famer Shannon Sharpe ripped Memphis Grizzlies All-Star Ja Morant over his latest incident involving a handgun. Sharpe initially said he was disappointed in Morant for his actions after he was previously suspended for a separate gun incident while intoxicated at a Denver-area nightclub. That incident led to an eight-game suspension by the league.

Abandon all hope, ye who merge spreadsheet cells! Last week, at its annual I/O conference, Google spent hours detailing how large language models would help the knowledge workers of the world unload their busywork onto a legion of eager, capable neural networks. The company will soon introduce AI functions into programs such as Gmail, Google Sheets, and Google Slides that will allow users to type simple commands and receive complex outputs: entire email compositions, for example, or auto-generated tables. The future that Google is promising feels familiar--it's all about heightened convenience and one-click efficiency--and I hate it. Workplace AI feels like the purest distillation of a corrosive ideology that demands frictionless productivity from workers: The easier our labor becomes, the more of it we can do, and the more of it we'll be expected to do.

Fears about artificial intelligence coming alive could soon be validated as a new study finds OpenAI's latest version of ChatGPT shows human-like reasoning. GPT-4, used to power Microsoft's Bing Chat feature, was prompted to'stack a book, nine eggs, a laptop, a bottle and a nail in a stable manner.' The system arranged the items so the eggs would not break, detailing how each should be placed on the other - starting with the book and ending with the nail. It also commented on arranging the items so the eggs do not crack - something only humans could fully understand. Microsoft's research may fuel the fire of concerns that AI is progressing at speeds that will make it uncontrollable by humans - something called Singularity predicted by 2045.

On Tuesday, lawmakers, A.I. experts, and the guy chiefly responsible for ChatGPT gathered in the same room to swap analogies for just how dramatically A.I. is about to change our lives. The invention of the internet, the cell phone, and airplanes all made the list. For a Senate Judiciary Committee hearing ostensibly concerned with the dangers A.I. might pose to the world, everyone seemed to get along quite well. At one point Sen. John Kennedy of Louisiana asked Sam Altman, the CEO of ChatGPT maker OpenAI, if he could recommend some people to oversee a new agency to oversee A.I.--that is, to pick his own regulators. Then again, Altman was doing an exceptional job projecting a self-critical persona.

As tedious as the incessant requests are for longer and harder-to-remember passwords, experts say there's good reason for the nuisance. It's gotten easier and easier for hackers to guess your password as computer processing speeds have gotten faster. With sprawling cloud-based computer power now available for rent to anyone -- and massive supercomputers out there, like the system that trained ChatGPT -- cyber security firm Hive Systems says that a truly professional hacker could access your secrets almost instantly. The company has produced a new table showing just how safe or vulnerable your password is, based on its character count and the diversity of characters you've used. They say you'll need a fully random password, that's at least 12-characters long, with a mixture of numbers, special symbols, upper- and lowercase letters, if you want to keep even just an amateur hacker out of your account, thanks to the power of today's consumer desktop tech.

Since the tech industry began its love affair with machine learning about a decade ago, US lawmakers have chattered about the potential need for regulation to rein in the technology. No proposal to regulate corporate AI projects has got close to becoming law--but OpenAI's release of ChatGPT in November has convinced some senators there is now an urgent need to do something to protect people's rights against the potential harms of AI technology. At a hearing held by a Senate Judiciary subcommittee yesterday attendees heard a terrifying laundry list of ways artificial intelligence can harm people and democracy. Senators from both parties spoke in support of the idea of creating a new arm of the US government dedicated to regulating AI. The idea even got the backing of Sam Altman, CEO of OpenAI.