Open Science Framework
PulseAugur coverage of Open Science Framework — every cluster mentioning Open Science Framework across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Gemini brand mentions show durability over 20 days in pre-registered test
A pre-registered study tested the durability of brand mentions in Gemini's responses over a 20-day period. The research found that forecasts based on historical mention rates performed well, outperforming simpler predic…
-
Fiorillo v0.5 open model released for randomized trial analysis
A new open-source model named Fiorillo v0.5 has been released, designed to answer questions about randomized trials with probabilistic outputs. This model, based on Qwen3-4B-Base with added adapters and a decision head,…
-
New PIMMUR principles reveal LLM social simulations may capture model bias, not human behavior
A new paper published on arXiv details the PIMMUR principles, a framework designed to ensure the validity of collective behavior simulations conducted using large language models (LLMs). Researchers audited 576 studies …
-
AI's statistical approximation falls short of general intelligence, study claims
A new preprint argues that current statistical approximation methods in AI do not equate to general intelligence. The paper suggests that while AI can excel at pattern recognition and prediction, it lacks the deeper und…
-
AI to streamline variant review across Asia, connecting evidence and experts
A new paper proposes using AI to facilitate variant review across Asia, focusing on reducing friction between distributed evidence and expert judgment rather than autonomous classification. The proposed workflow connect…
-
Transformer layers recode syntax towards canonical forms, study finds
A new research paper explores how Transformer layers process syntactic information, specifically focusing on Modern Greek. The study found that while early and middle layers can decode syntax, later layers appear to rec…
-
Study: Em-dash frequency in US congressional press releases rises with LLM adoption
A recent study analyzed over 146,000 U.S. congressional press releases from 2021 to 2025 to detect stylistic traces of large language model (LLM) usage. Researchers observed a significant increase in the frequency of un…
-
New Database Simplifies Search for Aesthetics Research Datasets
Researchers have developed DODA, a new web application designed to streamline the search for image datasets relevant to aesthetics research. This tool addresses the challenge of fragmented dataset sharing across platfor…
-
Unified Digital Organism OS Strategy Detailed Across Platforms
Primordial Calculus has detailed a "Digital Organism Operating System" and its associated "Substrate Strategy" across multiple platforms including Mastodon, X, Hugging Face, OSF, GitHub, and YouTube. This system is desc…
-
New OSF foundation models advance sleep physiology research
Researchers have introduced OSF, a new family of foundation models for sleep physiology, designed to address the heterogeneity in polysomnography data. Through extensive pre-training on a large corpus of sleep recording…
-
Study: Em-dash usage surged in scientific preprints post-LLM adoption
A recent study analyzed preprints from medRxiv and found a significant increase in the use of em-dashes following the widespread adoption of large language models (LLMs). The prevalence of em-dashes in discussion sectio…
-
New method enhances Chinese legislative conflict classification
Researchers have developed TypedCSIP, a novel counterfactual pretraining method designed for Chinese legislative conflict classification. This method leverages expert-written revisions as counterfactual supervision to t…
-
Study finds 3-9B LLMs fail verbal confidence tests, impacting uncertainty estimates
A new study examined the verbal confidence of seven instruction-tuned, open-weight large language models (LLMs) with 3-9 billion parameters. Researchers found that these models failed to meet minimal validity criteria f…