18062 research outputs found
Sort by
Examining the Molecular Mechanisms of Glucagon-like Peptide-1 Receptor Agonists in Cancer Cell Biology
No full textGlucagon-like peptide-1 receptor agonists (GLP-1RAs) are synthetic analogs of glucagon-like peptide-1 (GLP-1) used to treat obesity and diabetes by reducing blood glucose levels and appetite. While early rodent studies suggested a link between GLP-1RAs and thyroid cancer pathogenesis, evidence in humans remains inconclusive, with randomized controlled trials not supporting that link. Due to the rapid, widespread use of these drugs, concerns have expanded to other obesity-associated cancers, with conflicting findings on their role in cancer progression. With this contradictory evidence in a novel intersection of obesity medicine and oncology, this literature review aims to summarize current understandings between GLP-1-RAs and cancers to address these worries. Recent studies indicate that GLP-1RAs may regulate signaling pathways associated with cancer, including cAMP-mediated, PI3K/Akt/mTOR, AMPK, and NF-κB signaling, with pro- and anti-cancer effects depending on the cancer type. Preclinical and retrospective clinical studies have suggested that GLP-1RA treatment may alleviate liver, colorectal, and pancreatic cancer. However, research on thyroid, breast, and kidney cancers remains inconclusive, with some studies indicating cancer-promoting effects while others suggest no effect or even anti-cancer effects. Patients currently being treated should follow their physician’s recommendations; yet more research is needed to fully understand the molecular mechanisms of GLP-1RAs in cancer to provide clear guidance
Transporting HBCU Students to Morocco through Films: Action research and reflection on a piloted course
No full textSuccessful Career Construction in the Fourth Industrial Revolution: Young Graduates in the UAE
No full textThe Impact of Short-Term Study Abroad (STSA) Program on Intercultural Competence of Students at a Frontier/Rural Campus
No full text
Let it Flow: Information Exchange in Video Conferences versus Face-to-Face Meetings
No full textWhen the COVID-19 pandemic hit, policymakers faced a seemingly difficult choice. On the one hand, health considerations required imposing restrictions on face-to-face meetings. On the other, intuition suggested that switching to video conferencing might lead to information loss. As the pandemic progressed, in-person meetings largely turned digital, including court hearings, lawyer-client consultations, board meetings, and more. But did this turn actually cause an information loss?
Figuring out whether information is lost in video conferences is pivotal not only as a reflection on the pandemic but also to determine how to move forward in a post-pandemic world. In particular, identifying whether information is lost could answer the question, should policymakers permit video conferencing?
Briefly before the pandemic erupted, we conducted a lab experiment that focused precisely on this issue, foreshadowing its relevance. Subjects were asked to solve a riddle, which could only be solved correctly by exchanging information with other subjects. Our main research question was whether the medium used to communicate—video conference or face-to-face—mattered for information flow. Contrary to our expectations, we found no significant difference in information flow between the two mediums. However, we did observe a perception gap: those who communicated face-to-face were more likely to perceive themselves as giving away useful information. Yet this perception was entirely subjective, as face-to-face meetings did not, in fact, yield better performance.
Our findings entail two key implications. First, actors in the legal sphere, such as policymakers, lawyers, judges, shareholders, and board members, should be potentially (and counter-intuitively) less concerned about information loss in video conferencing. Second, policymakers should be aware that resistance to video conferences might stem from a biased perception of differences in information flow between the mediums. To maintain data-driven decisions, we also stress the need to collect additional evidence that accounts for experience with video conferencing during and after the pandemic
Gilead: Municipal Liability for Punitive Damages Under the Fair Housing Act
No full textThe 1968 Fair Housing Act (“FHA”) has always been understood to apply to local governments, which have proved to be among the most frequent and significant violators of this law, especially in their opposition to housing of particular value to racial minorities and persons with disabilities. Yet not until the Second Circuit’s decision last year in Gilead Community Services, Inc. v. Town of Cromwell did an appellate court approve an FHA-based punitive-damage award against a municipality. Before Gilead, district courts had generally blocked such awards, applying § 1983’s immunities to protect local governments and their officials from the FHA’s full set of remedies. In rejecting this approach, Gilead charts a new course, offering a potential breakthrough for deterring municipal housing discrimination.
This Article reviews the issues raised in Gilead and the potential impact of that decision. After providing the relevant background and analyzing the Second Circuit’s decision, the Article identifies the various types of FHA claims that have been brought against local governments. It then discusses other key issues— including whether Gilead’s endorsement of punitive damages in these cases might also extend to claims of non-intentional discrimination and whether Gilead might open the way for stripping local officials of their § 1983 immunities in FHA cases— that must be considered if Gilead’s promise of a more robust remedial arsenal for the FHA is to be fulfilled
Multimodal Benchmarking for NCAA Basketball
No full textWe present the first multimodal, multitask benchmark for NCAA basketball, synthesizing structured statistical features with large language model (LLM)-generated game summaries across 19,739 games spanning four NCAA Division I seasons (2021--2025). We evaluate three model families---XGBoost, deep neural networks, and Transformers---under tabular-only and early-fusion settings to measure the impact of LLM-derived textual embeddings. To assess practical utility, we simulate fixed-stake and Kelly criterion-based betting strategies using historical bookmaker odds, analyzing both profitability and downside risk via Monte Carlo simulation. Our results show that XGBoost with early-fusion achieves the highest return on investment and the lowest risk of loss. This work is, to our knowledge, the first to integrate LLM-generated narrative data with structured inputs for calibrated forecasting in sports, offering a reproducible benchmark for multimodal decision-making under uncertainty
