arXiv Computer Science @arxiv_cs@qoto.org

Enhancing Safety of Large Language Models via Embedding Space Separation https://arxiv.org/abs/2603.20206 #cs.CL #cs.AI

RedacBench: Can AI Erase Your Secrets? https://arxiv.org/abs/2603.20208 #cs.CL #cs.AI #cs.CR

Children's Intelligence Tests Pose Challenges for MLLMs? KidGym: A 2D Grid-Based Reasoning Benchmark for MLLMs https://arxiv.org/abs/2603.20209 #cs.CL #cs.AI

CRoCoDiL: Continuous and Robust Conditioned Diffusion for Language https://arxiv.org/abs/2603.20210 #cs.CL #cs.AI

Exploring Teacher-Chatbot Interaction and Affect in Block-Based Programming https://arxiv.org/abs/2603.20211 #cs.CY #cs.AI

Engineering-Oriented Symbolic Regression: LLMs as Physics Agents for Discovery of Simulation-Ready Constitutive Laws https://arxiv.org/abs/2603.19241 #physics.comp-ph #physics.app-ph #cs.CE #cs.SC

The IJCNN 2025 Review Process https://arxiv.org/abs/2603.19244 #cs.DL #cs.LG

How international are international computing conferences? -- An exploration with systems research conferences https://arxiv.org/abs/2603.19245 #cs.OH

Speed and impact of team science during urgent societal events https://arxiv.org/abs/2603.19246 #cs.DL

When Prompt Optimization Becomes Jailbreaking: Adaptive Red-Teaming of Large Language Models https://arxiv.org/abs/2603.19247 #cs.CL #cs.AI

Negative Sampling Techniques in Information Retrieval: A Survey https://arxiv.org/abs/2603.18005 #cs.IR

Do Large Language Models Possess a Theory of Mind? A Comparative Evaluation Using the Strange Stories Paradigm https://arxiv.org/abs/2603.18007 #cs.CL #cs.AI

TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots https://arxiv.org/abs/2603.18008 #cs.CL #cs.AI #cs.CY

How Confident Is the First Token? An Uncertainty-Calibrated Prompt Optimization Framework for Large Language Model Classification and Understanding https://arxiv.org/abs/2603.18009 #cs.CL #cs.AI

Agentic Framework for Political Biography Extraction https://arxiv.org/abs/2603.18010 #cs.CL #cs.AI #cs.CY

Controllable Evidence Selection in Retrieval-Augmented Question Answering via Deterministic Utility Gating https://arxiv.org/abs/2603.18011 #cs.CL #cs.IR

DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation https://arxiv.org/abs/2603.18012 #cs.CL #cs.AI #cs.IR

Learned but Not Expressed: Capability-Expression Dissociation in Large Language Models https://arxiv.org/abs/2603.18013 #cs.CL

Real-Time Trustworthiness Scoring for LLM Structured Outputs and Data Extraction https://arxiv.org/abs/2603.18014 #cs.CL #cs.LG

Beyond Accuracy: An Explainability-Driven Analysis of Harmful Content Detection https://arxiv.org/abs/2603.18015 #cs.CL #cs.AI