Large Language Model-Based Generation of Discharge Summaries

French AI startup Mistral has launched the Devstral 2 coding model, which includes a laptop-friendly version optimized for software engineering tasks. This release follows the introduction of the Mistral 3 LLM family, aimed at enhancing local hardware capabilities for developers.

A new online structured pruning technique called Token Filtering has been introduced for large language models (LLMs), allowing pruning decisions to be made during inference without the need for calibration data. This method measures token redundancy through joint key-value similarity, effectively reducing inference costs while maintaining essential information. The approach also includes a variance-aware fusion strategy to ensure important tokens are preserved even with high pruning ratios.

The introduction of Graph-Regularized Sparse Autoencoders (GSAEs) aims to enhance the safety of large language models (LLMs) by addressing their vulnerabilities to adversarial prompts and jailbreak attacks. GSAEs extend traditional sparse autoencoders by incorporating a Laplacian smoothness penalty, allowing for the recovery of distributed safety representations across multiple features rather than isolating them in a single latent dimension.

A new method called Depth-Wise Activation Steering has been introduced to enhance the honesty of large language models (LLMs) like LLaMA, Qwen, and Mistral. This training-free approach utilizes a Gaussian schedule to improve the models' ability to report truthfully, addressing the issue of models asserting falsehoods despite having the correct information internally.

Recent research indicates that large language models (LLMs) demonstrate biases in evaluation tasks, particularly favoring self-generated content. However, a study exploring retrieval-augmented generation (RAG) frameworks found no significant self-preference effect, suggesting that LLMs can evaluate factual content more impartially than previously thought.

Large language models (LLMs) have revolutionized automated software development, enabling the conversion of natural language into functional code, as highlighted in a comprehensive survey on code intelligence. This evolution is exemplified by tools like Github Copilot and Claude Code, which have significantly improved coding success rates on benchmarks like HumanEval.

Recent advancements in natural language processing have led to the development of Confounding-Aware Token Rationalization (CATR), a framework designed to improve causal effect estimation by selecting a sparse subset of text tokens. This approach addresses challenges posed by high-dimensional text data, particularly the violation of the positivity assumption in treatment effect estimation.

A recent study highlights the potential of using GPT-4, along with ICD-10 and clinical ontology knowledge graphs, to enhance the generation of clinical notes from electronic health records (EHRs). This approach leverages Chain-of-Thought prompting to improve the accuracy and efficiency of note-taking by physicians, addressing the time-consuming nature of manual entries.