LLMs detect falsified ISO/IEC/IEEE 24765 SE definitions well but also reject many correct ones, indicating systematic rejection bias rather than genuine discriminative understanding of SE terminology.
GLUE: A multi-task benchmark and analysis platform for natural language understanding
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
Benchmarking 25 LLMs on Raspberry Pi hardware shows Granite4 Tiny Hybrid (7B) balances 2.5 tokens/s, 0.90 tokens/J, and 54.6% MMLU while teaching effectiveness does not require high general knowledge scores.
FDA-Opt unifies and improves upon FedOpt and FDA for communication-efficient federated fine-tuning of language models on NLP tasks, outperforming optimized FedOpt baselines.
citing papers explorer
-
Large Language Models Have Unreliable Understanding of Software Engineering Terminology
LLMs detect falsified ISO/IEC/IEEE 24765 SE definitions well but also reject many correct ones, indicating systematic rejection bias rather than genuine discriminative understanding of SE terminology.
-
Benchmarking Local Language Models for Social Robots using Edge Devices
Benchmarking 25 LLMs on Raspberry Pi hardware shows Granite4 Tiny Hybrid (7B) balances 2.5 tokens/s, 0.90 tokens/J, and 54.6% MMLU while teaching effectiveness does not require high general knowledge scores.
-
Communication-Efficient Federated Fine-Tuning
FDA-Opt unifies and improves upon FedOpt and FDA for communication-efficient federated fine-tuning of language models on NLP tasks, outperforming optimized FedOpt baselines.