small language model

Quality, safety, and readability of consumer-facing generative AI responses to end-of-life questions relevant to surrogate decision-makers for older adults: a benchmark evaluation

As the population ages and the prevalence of multiple coexisting conditions increases, end-of-life decision-making for older adults is becoming increasingly complex in clinical practice. limited access to relevant medical information and differences in medical knowledge between clinicians and surrogate decision-makers....

Generating Explainable Domain Models via Attention and Confidence

LLM-Driven Verilog Generation and Verification for RISC-V Processor Design

Large language models (LLMs) are increasingly used in Electronic Design Automation (EDA) to write hardware description code. This paper reviews how LLMs generate and verify Verilog for a RISC-V processor datapath, a core topic in Computer Architecture and Organization. The review is built around the AI-driven logic syn...

Responsible use of large language models in digital health: an equity-first governance framework

Large language models (LLMs) are being rapidly deployed for digital health, yet their equity impacts remain poorly characterized. We present an equity-first audit framework anchored in the first nations mental wellness continuum framework (FNMWCF) and apply it to three open-source LLMs (LLaMA-3.2:latest, Mistral-7B, an...

MIND-T1D: study protocol for a multidimensional neurocognitive and psychological assessment of Indian children and young adults with type 1 diabetes.

The MIND-T1D protocol is described, a culturally sensitive, clinically feasible, family-centred assessment framework for Indian children and young adults with T1D that addresses language-related barriers and educational variability by incorporating neuropsychological assessments and integrating child, parent and biolog...

Single-Pass Uncertainty Heads for Claim-Level Hallucination Detection in Persian Medical Language Models

Hallucination detection is particularly important for medical language models, but repeated-sampling approaches are computationally expensive. A faster alternative is a single-pass uncertainty head that predicts hallucination risk from a frozen generator's internal signals. Existing uncertainty heads consume backbone-s...

Learning When to Commit from Partial Speech for End-to-End Simultaneous Speech Translation

Simultaneous speech translation must emit useful target text before the source is complete while preserving every committed token. We adapt a full-utterance speech language model using prefix supervision derived from its own complete- and partial-waveform translations, requiring neither transcripts nor human translatio...

FALCON: A Model and Dataset Agnostic Framework for Synthetic Data Generation for NL2SQL Pairs

Relational databases are among the most widely deployed forms of structured knowledge, and natural language access to them requires grounding language onto schema entities and relations while handling the ambiguity inherent in how people phrase requests. Existing synthetic NL-to-SQL data generation methods largely igno...

HyperThink: Text-to-Parameter Hypernetworks for Efficient Reasoning

Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LLMs), but they introduce high inference-time overhead, with latency dominated by sequential decoding. We propose HyperThink, a text-to-parameter approach that amortizes this reasoning computation into a s...

LESSER: Post-Training Data Selection with Output-Layer Gradients

The choice of post-training data for large language models substantially affects downstream performance. Gradient-based data selection is a popular approach that ranks training data by how well their gradients align with those of a small validation set. However, ranking with full-parameter gradients requires an expensi...

Exact Memory-Time Optimization for Prefix-Cached Language Model Serving

Retaining language-model prefix states trades recomputation against storage time. Optimizing each cached block independently can overcount savings: a resident block is usable only when the required preceding prefix is also available. We introduce Prefix-Certificate Retention (PCR), an exact finite-trace formulation for...

Online Verification of Language Model Responses Under Cost Constraints

As large language models are increasingly deployed for multi-step reasoning, verifying the correctness of their outputs has become essential for maintaining reliability at scale. Verifying the correctness of large language model outputs is often done by querying a costly ground-truth oracle, which is impractical to inv...