MedProb is a lightweight probing framework that predicts multiple-choice medical visual question answering (Med‑VQA) answers directly from frozen vision‑language model (VLM) representations, avoiding free‑text generation. On datasets such as PATH‑VQA, SLAKE, and VQA‑RAD, MedProb extracts more answer‑relevant signal than prompting and outperforms both medical VLMs and agentic systems. The approach also narrows the performance gap between small and large models, shows that medical adaptation does not consistently improve linear decodability, and reveals positional biases in both prompting and generation.
By Erfan Nourbakhsh, Ke Yang, Anthony Rios
arXiv:2610.34871v1 Announce Type: new
Abstract: Objective: The rapid advancement of large language models (LLMs) in healthcare has opened new possibilities for AI-assisted clinical decision-making; h...
By Asad Gohar, Md Ariful Islam Mozumder, Moon-Il Joo, Hee-Cheol Kim
The paper introduces a method for text simplification that uses a Large Language Model-as-a-Judge (LLM‑Judge) to automatically generate policy‑aligned training data, eliminating the need for costly human annotation or parallel corpora. This approach allows simplification models to adapt to various policies—such as lexical‑focused simplification or overall rewriting that trades detail for simplicity—at both sentence and document levels. Experiments show that small‑scale open‑source LLMs can match or surpass GPT‑5, with consistent improvements across model families and sizes.
By Xuanxin Wu, Yuki Arase, Masaaki Nagata
arXiv:2509. 21530v2 Announce Type: replace Abstract: Data augmentation is a widely used strategy to improve model robustness and generalization by enriching training datasets with synthetic examples.
By Dongkyu Cho, Miao Zhang, Rumi Chunara
The paper PRiME investigates how large language models (LLMs) can perform pragmatic reasoning in the mental health domain. It introduces three datasets and tasks, evaluates 12 LLMs—including the specialized MentaLLaMA model—and finds that performance is comparable to human evaluation, with Mistral‑7B achieving the best results. The study also examines error asymmetry and uses first‑order attention rollout to analyze MentaLLaMA‑7B’s architectural behavior, urging careful analysis for high‑stakes applications.
By Sneha Oram, Tejomay Kishor Padole, Pushpak Bhattacharyya, Suyash P. Awate
arXiv:2608.24327v2 Announce Type: replace
Abstract: With the advent of Large Language Models and its instruction following capabilities a promising application is the task of summarization. Within th...
By Enes Yavuz Ugan, Fabian Retkowski, Yuka Ko, Thai-Binh Nguyen, Maike Z\"ufle, Jan Niehues, Alexander Waibel