CORTEXA
← Browse
arxivcs.LGcs.AIcs.CL2026-07-01

On the Utility and Factual Reliability of Pruned Mixture-of-Experts Models in the Biomedical Domain

Atsuki Yamaguchi, Szymon Palucha, Léo Bijar, Aline Villavicencio, Nikolaos Aletras

Mixture-of-Experts (MoE) models offer inference speedups via selective activation but impose substantial memory requirements because the whole network must remain loaded. Structured expert pruning is a practical approach for reducing deployment costs in resource-constrained settings. However, prior studies primarily evaluate benchmark utility, leaving the effect of pruning on factual reliability underexplored, particularly in high-stakes domains such as biomedicine. In this paper, we investigate how domain-specific expert pruning affects both utility and reliability. We assess four MoE models, six pruning methods, and multiple pruning ratios across generation and classification tasks under in-domain (biomedical) and cross-domain settings. Results reveal that moderate pruning preserves in-domain utility without immediate reliability decline, although hallucination risks increase at extreme pruning ratios. When shifting to the general domain, both utility and reliability degrade rapidly. These findings indicate that safe compression depends heavily on the task and domain. Evaluating pruned MoE models solely on utility is inadequate for high-stakes deployment without reliability assessment.

View free PDFSource page

Related papers

arxivcs.CVcs.AIcs.CLcs.LGcs.MM2026-07-01

ESC: Emotional Self-Correction for Reliable Vision-Language Models

Tien-Huy Nguyen, Minh-Nhat Nguyen, Nguyen Nhat Huy, Hung Viet Nguyen, Huy Nguyen Minh Nhat, Thanh-Huy Nguyen, et al.

Vision-language models (VLMs) have achieved strong performance across diverse multimodal tasks, yet they remain vulnerable to unreliable reasoning. Existing self-correction methods mitigate these issues but typically rely on post-training or carefully engineered feedback, incurri…

View free PDFSource page
arxivcs.ARcs.AIcs.CLcs.DCcs.LGcs.PF2026-07-21

BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators

Fabian Waschkowski, Prabod Rathnayaka, Lukas Wesemann

Apple's M5 generation introduces a redesigned GPU architecture in which every core carries a dedicated Neural Accelerator: on-die matrix units exposed through the Metal~4 tensor API. We show that BaseRT, our native Metal inference runtime for large language models on Apple Silico…

View free PDFSource page
arxivcs.LGcs.AIcs.CL2026-07-01

CausalMix: Data Mixture as Causal Inference for Language Model Training

Zinan Tang, Yukun Zhang, Shaomian Zheng, Zhuoshi Pan, Qizhi Pei, Dingnan Jin, et al.

In Large Language Model (LLM) training, data mixing plays a pivotal role in determining model performance. Recent methods optimize mixture weights via proxy models, but they rely on the assumption of static data distributions. As a result, when the underlying data pool shifts, th…

View free PDFSource page
arxivcs.LGcs.AIcs.CL2026-06-30

Reliability Scales Inversely: Bigger Models Compound Mistakes Faster via a Hidden Auto-Regressive Risk Regime

Kushal Chakrabarti

As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability. The knowledge-gap account - more data, retrieval, or scale - misses an auto-regressive risk residual that scale sharpens: the model commits to a low-probability token…

View free PDFSource page
arxivcs.CLcs.AIcs.LG2026-07-16

In-Place Tokenizer Expansion for Pre-trained LLMs

Jimmy T. H. Smith, Tarek Dakhran, Alberto Cabrera, Simon S. Lee, Paul Pak, Aditya Tadimeti, et al.

A tokenizer fixed at the start of pre-training allocates vocabulary in proportion to the pre-training corpus, reflecting the deployment priorities at that time. When those priorities shift, languages added later are split into many more tokens per word, which can raise latency, c…

View free PDFSource page
arxivcs.LGcs.AIcs.CLcs.CV2026-07-05

DynaVieW: Schema-Guided World Modeling for Understanding Hierarchical Visual Dynamics

Silin Gao, Hao Zhao, Zeming Chen, Sepideh Mamooler, Antara Raaghavi Bhattacharya, Qiyu Wu, et al.

Multimodal LLMs struggle to systematically model the temporal evolution of visual scenes in videos or multi-image sequences. Such inputs require models to predict or simulate multiple levels of dynamic constituents, such as actions taken in the visual sequence, and the associated…

View free PDFSource page