Evaluating LLM Trust in Conflicting Software Artifacts
Discover how TRACE framework measures LLM trust allocation across code, docs, and tests to improve reliability in software engineering.
Agile Story-Point Estimation: Is RAG More Effective?
Explore how Retrieval Augmented Generation (RAG) can improve Agile story-point estimation by saving time and boosting accuracy in sprint planning.
MetaSAEs: Enhancing Atomic Sparse Autoencoder Latents
Discover how MetaSAEs use joint training with decomposability penalties to create more atomic, interpretable sparse autoencoder latents for safer AI models...
Optimize Marine Species Classification with Circuit Duplication
Enhance marine species classification using circuit duplication in frozen visual transformers—boost accuracy without fine-tuning or extra training.
AEGIS: Efficient Multi-GPU Scaling for Encrypted Transformer Inference
Discover AEGIS, a hybrid parallelism framework that scales long-sequence homomorphic encrypted Transformer inference efficiently on multi-GPU systems.