arXiv Computer Science @arxiv_cs@qoto.org

Inclusion-of-Thoughts: Mitigating Preference Instability via Purifying the Decision Space https://arxiv.org/abs/2604.04944 #cs.CL #cs.AI

Sparse Autoencoders as a Steering Basis for Phase Synchronization in Graph-Based CFD Surrogates https://arxiv.org/abs/2604.04946 #physics.comp-ph #cs.CE #cs.LG

IC3-Evolve: Proof-/Witness-Gated Offline LLM-Driven Heuristic Evolution for IC3 Hardware Model Checking https://arxiv.org/abs/2604.03232 #cs.AI

Integrating Artificial Intelligence, Physics, and Internet of Things: A Framework for Cultural Heritage Conservation https://arxiv.org/abs/2604.03233 #math.NA #cs.LG #cs.NA

Structural Segmentation of the Minimum Set Cover Problem: Exploiting Universe Decomposability for Metaheuristic Optimization https://arxiv.org/abs/2604.03234 #math.OC #cs.AI

Toward a Universal Color Naming System: A Clustering-Based Approach using Multisource Data https://arxiv.org/abs/2604.03235 #cs.HC #cs.AI #cs.CV

BLADE: Better Language Answers through Dialogue and Explanations https://arxiv.org/abs/2604.03236 #cs.HC #cs.CL

The Persuasion Paradox: When LLM Explanations Fail to Improve Human-AI Team Performance https://arxiv.org/abs/2604.03237 #cs.HC #cs.AI #cs.CL

Measuring Human Preferences in RLHF is a Social Science Problem https://arxiv.org/abs/2604.03238 #cs.HC

To Throw a Stone with Six Birds: On Agents and Agenthood https://arxiv.org/abs/2604.03239 #cs.AI

Scaling DPPs for RAG: Density Meets Diversity https://arxiv.org/abs/2604.03240 #cs.LG #cs.AI #cs.CL

Incidental Interaction: Technology to Support Elder Strength Training through Everyday Movements https://arxiv.org/abs/2604.03241 #cs.HC

Not All Denoising Steps Are Equal: Model Scheduling for Faster Masked Diffusion Language Models https://arxiv.org/abs/2604.02340 #cs.LG #cs.CL

LLM Reasoning with Process Rewards for Outcome-Guided Steps https://arxiv.org/abs/2604.02341 #cs.LG #cs.AI

Homophily-aware Supervised Contrastive Counterfactual Augmented Fair Graph Neural Network https://arxiv.org/abs/2604.02342 #cs.LG

Haiku to Opus in Just 10 bits: LLMs Unlock Massive Compression Gains https://arxiv.org/abs/2604.02343 #math.IT #cs.LG #cs.AI #cs.IT

Characterizing WebGPU Dispatch Overhead for LLM Inference Across Four GPU Vendors, Three Backends, and Three Browsers https://arxiv.org/abs/2604.02344 #cs.LG #cs.DC #cs.PF

DOne: Decoupling Structure and Rendering for High-Fidelity Design-to-Code Generation https://arxiv.org/abs/2604.01226 #cs.CV #cs.SE

Logic-Gated Time-Shared Feedforward Networks for Alternating Finite Automata: Exact Simulation and Learnability https://arxiv.org/abs/2604.01228 #cs.FL #cs.AI

CLPIPS: A Personalized Metric for AI-Generated Image Similarity https://arxiv.org/abs/2604.01234 #eess.IV #cs.CV #cs.AI