Large Language Models

Google’s Gemini AI can answer your questions with 3D models and simulations

Google's latest upgrade for Gemini will allow the chatbot to generate interactive 3D models and simulations in response to your questions...

Moody’s Integrates AI Agents With Anthropic’s Claude

AI on the couch: Anthropic gives Claude 20 hours of psychiatry

[2603.03330] Certainty robustness: Evaluating LLM stability under self-challenging prompts

Abstract page for arXiv paper 2603.03330: Certainty robustness: Evaluating LLM stability under self-challenging prompts

[2603.03329] AutoHarness: improving LLM agents by automatically synthesizing a code harness

Abstract page for arXiv paper 2603.03329: AutoHarness: improving LLM agents by automatically synthesizing a code harness

[2603.03328] StructLens: A Structural Lens for Language Models via Maximum Spanning Trees

Abstract page for arXiv paper 2603.03328: StructLens: A Structural Lens for Language Models via Maximum Spanning Trees

[2603.03326] Controllable and explainable personality sliders for LLMs at inference time

Abstract page for arXiv paper 2603.03326: Controllable and explainable personality sliders for LLMs at inference time

[2603.03325] IntPro: A Proxy Agent for Context-Aware Intent Understanding via Retrieval-conditioned Inference

Abstract page for arXiv paper 2603.03325: IntPro: A Proxy Agent for Context-Aware Intent Understanding via Retrieval-conditioned Inference

[2603.03324] Controlling Chat Style in Language Models via Single-Direction Editing

Abstract page for arXiv paper 2603.03324: Controlling Chat Style in Language Models via Single-Direction Editing

[2603.03323] Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement

Abstract page for arXiv paper 2603.03323: Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement

[2603.03322] Can Large Language Models Derive New Knowledge? A Dynamic Benchmark for Biological Knowledge Discovery

Abstract page for arXiv paper 2603.03322: Can Large Language Models Derive New Knowledge? A Dynamic Benchmark for Biological Knowledge Di...

[2603.03321] DIALEVAL: Automated Type-Theoretic Evaluation of LLM Instruction Following

Abstract page for arXiv paper 2603.03321: DIALEVAL: Automated Type-Theoretic Evaluation of LLM Instruction Following

[2603.03389] Towards Improved Sentence Representations using Token Graphs

Abstract page for arXiv paper 2603.03389: Towards Improved Sentence Representations using Token Graphs

[2603.03320] From We to Me: Theory Informed Narrative Shift with Abductive Reasoning

Abstract page for arXiv paper 2603.03320: From We to Me: Theory Informed Narrative Shift with Abductive Reasoning

[2603.03319] Automated Concept Discovery for LLM-as-a-Judge Preference Analysis

Abstract page for arXiv paper 2603.03319: Automated Concept Discovery for LLM-as-a-Judge Preference Analysis

[2603.03378] AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis

Abstract page for arXiv paper 2603.03378: AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis

[2603.03318] Quantum-Inspired Self-Attention in a Large Language Model

Abstract page for arXiv paper 2603.03318: Quantum-Inspired Self-Attention in a Large Language Model

[2603.03314] Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO

Abstract page for arXiv paper 2603.03314: Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO

[2603.03313] How does fine-tuning improve sensorimotor representations in large language models?

Abstract page for arXiv paper 2603.03313: How does fine-tuning improve sensorimotor representations in large language models?

[2603.03308] Old Habits Die Hard: How Conversational History Geometrically Traps LLMs

Abstract page for arXiv paper 2603.03308: Old Habits Die Hard: How Conversational History Geometrically Traps LLMs

[2603.03306] Token-Oriented Object Notation vs JSON: A Benchmark of Plain and Constrained Decoding Generation

Abstract page for arXiv paper 2603.03306: Token-Oriented Object Notation vs JSON: A Benchmark of Plain and Constrained Decoding Generation

[2603.03305] Draft-Conditioned Constrained Decoding for Structured Generation in LLMs

Abstract page for arXiv paper 2603.03305: Draft-Conditioned Constrained Decoding for Structured Generation in LLMs

[2603.03303] HumanLM: Simulating Users with State Alignment Beats Response Imitation

Abstract page for arXiv paper 2603.03303: HumanLM: Simulating Users with State Alignment Beats Response Imitation