On the Identifiability of Steering Vectors in Large Language Models
arXiv:2602.06801v1 Announce Type: cross Abstract: Activation steering methods, such as persona vectors, are widely used to control large language model…
Emergent Low-Rank Training Dynamics in MLPs with Smooth Activations
arXiv:2602.06208v1 Announce Type: cross Abstract: Recent empirical evidence has demonstrated that the training dynamics of large-scale deep neural networks occur…
Learning to Discover at Test Time
arXiv:2601.16175v2 Announce Type: replace-cross Abstract: How can we use AI to discover a new state of the art for a…
BioLite U-Net: Edge-Deployable Semantic Segmentation for In Situ Bioprinting Monitoring
arXiv:2509.06690v2 Announce Type: replace-cross Abstract: Bioprinting is a rapidly advancing field that offers a transformative approach to fabricating tissue and…
Sounding Highlights: Dual-Pathway Audio Encoders for Audio-Visual Video Highlight Detection
arXiv:2602.03891v2 Announce Type: replace-cross Abstract: Audio-visual video highlight detection aims to automatically identify the most salient moments in videos by…
Addressing Corpus Knowledge Poisoning Attacks on RAG Using Sparse Attention
arXiv:2602.04711v2 Announce Type: replace-cross Abstract: Retrieval Augmented Generation (RAG) is a highly effective paradigm for keeping LLM-based responses up-to-date and…
CAViT — Channel-Aware Vision Transformer for Dynamic Feature Fusion
arXiv:2602.05598v1 Announce Type: cross Abstract: Vision Transformers (ViTs) have demonstrated strong performance across a range of computer vision tasks by…
Reflexis: Supporting Reflexivity and Rigor in Collaborative Qualitative Analysis through Design for Deliberation
arXiv:2601.15445v2 Announce Type: replace-cross Abstract: Reflexive Thematic Analysis (RTA) is a critical method for generating deep interpretive insights. Yet its…
DyTopo: Dynamic Topology Routing for Multi-Agent Reasoning via Semantic Matching
arXiv:2602.06039v1 Announce Type: new Abstract: Multi-agent systems built from prompted large language models can improve multi-round reasoning, yet most existing…