Large Language Models

8 free AI courses from Anthropic’s Claude platform with certificates

How is mythos mythos ? [D]

Hello, I’ve been seeing discussions about “Mythos AI” showing behaviors that seem far beyond simple text prediction—like accessing inform...

Claude developer hosts Christian leaders for AI summit

[2603.01454] VidDoS: Universal Denial-of-Service Attack on Video-based Large Language Models

Abstract page for arXiv paper 2603.01454: VidDoS: Universal Denial-of-Service Attack on Video-based Large Language Models

[2603.01438] Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents

Abstract page for arXiv paper 2603.01438: Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing...

[2603.01385] Toward Graph-Tokenizing Large Language Models with Reconstructive Graph Instruction Tuning

Abstract page for arXiv paper 2603.01385: Toward Graph-Tokenizing Large Language Models with Reconstructive Graph Instruction Tuning

[2603.01780] D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation

Abstract page for arXiv paper 2603.01780: D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation

[2603.01761] Modular Memory is the Key to Continual Learning Agents

Abstract page for arXiv paper 2603.01761: Modular Memory is the Key to Continual Learning Agents

[2603.01759] Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning

Abstract page for arXiv paper 2603.01759: Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning

[2603.01343] PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

Abstract page for arXiv paper 2603.01343: PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

[2603.01752] Causal Circuit Tracing Reveals Distinct Computational Architectures in Single-Cell Foundation Models: Inhibitory Dominance, Biological Coherence, and Cross-Model Convergence

Abstract page for arXiv paper 2603.01752: Causal Circuit Tracing Reveals Distinct Computational Architectures in Single-Cell Foundation M...

[2603.01331] MetaState: Persistent Working Memory for Discrete Diffusion Language Models

Abstract page for arXiv paper 2603.01331: MetaState: Persistent Working Memory for Discrete Diffusion Language Models

[2603.01692] Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

Abstract page for arXiv paper 2603.01692: Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

[2603.01254] LLM Self-Explanations Fail Semantic Invariance

Abstract page for arXiv paper 2603.01254: LLM Self-Explanations Fail Semantic Invariance

[2603.01252] Linking Knowledge to Care: Knowledge Graph-Augmented Medical Follow-Up Question Generation

Abstract page for arXiv paper 2603.01252: Linking Knowledge to Care: Knowledge Graph-Augmented Medical Follow-Up Question Generation

[2603.01589] SafeSci: Safety Evaluation of Large Language Models in Science Domains and Beyond

Abstract page for arXiv paper 2603.01589: SafeSci: Safety Evaluation of Large Language Models in Science Domains and Beyond

[2603.01246] Defensive Refusal Bias: How Safety Alignment Fails Cyber Defenders

Abstract page for arXiv paper 2603.01246: Defensive Refusal Bias: How Safety Alignment Fails Cyber Defenders

[2603.01239] Self-Anchoring Calibration Drift in Large Language Models: How Multi-Turn Conversations Reshape Model Confidence

Abstract page for arXiv paper 2603.01239: Self-Anchoring Calibration Drift in Large Language Models: How Multi-Turn Conversations Reshape...

[2603.01563] LFPO: Likelihood-Free Policy Optimization for Masked Diffusion Models

Abstract page for arXiv paper 2603.01563: LFPO: Likelihood-Free Policy Optimization for Masked Diffusion Models

[2603.01224] Monocular 3D Object Position Estimation with VLMs for Human-Robot Interaction

Abstract page for arXiv paper 2603.01224: Monocular 3D Object Position Estimation with VLMs for Human-Robot Interaction

[2603.01501] GAC: Stabilizing Asynchronous RL Training for LLMs via Gradient Alignment Control

Abstract page for arXiv paper 2603.01501: GAC: Stabilizing Asynchronous RL Training for LLMs via Gradient Alignment Control

[2603.01185] Token-level Data Selection for Safe LLM Fine-tuning

Abstract page for arXiv paper 2603.01185: Token-level Data Selection for Safe LLM Fine-tuning

[2603.01170] ATLAS: AI-Assisted Threat-to-Assertion Learning for System-on-Chip Security Verification

Abstract page for arXiv paper 2603.01170: ATLAS: AI-Assisted Threat-to-Assertion Learning for System-on-Chip Security Verification