small language model

PSL-128: A Multi-Signer Pakistan Sign Language Video Dataset (224x224 preprocessed release)

PSL-128 is a multi-signer video dataset for word-level isolated sign recognition in Pakistan Sign Language (PSL), comprising 9,068 clips across 128 sign classes contributed by 68 signers. It is, to the authors' knowledge, the first large-scale, multi-signer, publicly released PSL video dataset for word-level recognitio...

Latent Steering of Emotional Evidence with Sparse Autoencoders in Language Models

Police accountability data: United Kingdom, Ireland, United States and Australia

Derived, machine-readable datasets on police complaints, stop and search, use of force, police workforce strength and deaths following police contact, for the United Kingdom, Ireland, the United States and Australia. Every figure is computed from official open data by code and published with its source, its period and...

From universal provision to equitable access: city-level linguistic diversity and access to maternal and child health education among migrant mothers in China

Migration changes not only where mothers live, but also the linguistic and institutional environments through which they encounter health information, primary public health systems, and community support. Existing research has largely explained disparities in maternal and child health services among migrants in terms o...

Protein Language Model Embeddings Recover Evolutionary and Functional Organization Across the Ras Superfamily.

Analysis of a curated Ras superfamily dataset indicates that protein language model embeddings provide an effective alignment-free representation of Ras family and subfamily organization, and can highlight candidate sequence regions associated with functional specialization from sequence alone.

zmactep/patinae: Release v0.5.0

Patinae v0.5.0 Zawinski's law says that programs keep expanding until they can read email. Patinae started as a molecular viewer. It now has background tasks, Markdown output, and an AI plugin that can connect to MCP servers. We appear to be one mail server away from compliance. The largest change in 0.5.0 is underneat...

PSL-128: A Multi-Signer Pakistan Sign Language Video Dataset (224x224 preprocessed release)

PSL-128 is a multi-signer video dataset for word-level isolated sign recognition in Pakistan Sign Language (PSL), comprising 9,068 clips across 128 sign classes contributed by 68 signers. It is, to the authors' knowledge, the first large-scale, multi-signer, publicly released PSL video dataset for word-level recognitio...

Artificial Intelligence and Radiomics for Non-Invasive Prediction of Genomic Alterations in Solid Tumors: A Scoping Review

This scoping review aims to map the current extent, range, and nature of evidence on the use of artificial intelligence (AI), machine learning, deep learning, radiomics, and radiogenomics applied to medical imaging for the non-invasive prediction or characterization of genomic alterations and molecular phenotypes in pa...

CHAPTER 15 - On Judicial Impartiality and Private Affiliations

CHAPTER 15 --- Recension- CDJ Opinion 2026‑27 on Judicial Impartiality and Private Affiliations --- The Collège de déontologie des magistrats de l’ordre judiciaire (CDJ) issued Opinion 2026‑27 on 9 June 2026, clarifying when a judge’s membership in an irregular fraternity (and, by extension, other private associations)...

CHAPTER 15 - On Judicial Impartiality and Private Affiliations

CHAPTER 15 --- Recension- CDJ Opinion 2026‑27 on Judicial Impartiality and Private Affiliations --- The Collège de déontologie des magistrats de l’ordre judiciaire (CDJ) issued Opinion 2026‑27 on 9 June 2026, clarifying when a judge’s membership in an irregular fraternity (and, by extension, other private associations)...

zmactep/patinae: Release v0.5.0

Patinae v0.5.0 Zawinski's law says that programs keep expanding until they can read email. Patinae started as a molecular viewer. It now has background tasks, Markdown output, and an AI plugin that can connect to MCP servers. We appear to be one mail server away from compliance. The largest change in 0.5.0 is underneat...

quickthink: local-first inference control layer for small LLMs

quickthink is a local-first inference control layer that aims to give small, on-device language models more reliable structured output by inserting a compressed plan-then-answer scaffold with latency-aware routing. Its routing, plan-grammar, and validation logic run as plain-Python middleware (no extra model of its own...