Large Language Models
GPT, Claude, Gemini, and other LLMs
Top This Week
Llms
Anthropic launches Claude Managed Agents — composable APIs for shipping production AI agents 10x faster. Notion, Rakuten, Asana, and Sentry already in production.
Anthropic launches Claude Managed Agents in public beta — composable APIs for shipping production AI agents 10x faster Handles sandboxing...
Reddit - Artificial Intelligence
·
1 min
·
Llms
6 Months Using AI for Actual Work: What's Incredible, What's Overhyped, and What's Quietly Dangerous
Six months ago I committed to using AI tools for everything I possibly could in my work. Every day, every task, every workflow. Here's th...
Reddit - Artificial Intelligence
·
1 min
·
All Content
![[2603.02675] From Shallow to Deep: Pinning Semantic Intent via Causal GRPO](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
Llms
[2603.02675] From Shallow to Deep: Pinning Semantic Intent via Causal GRPO
Abstract page for arXiv paper 2603.02675: From Shallow to Deep: Pinning Semantic Intent via Causal GRPO
arXiv - Machine Learning
·
3 min
·
Llms
[2504.21023] Param$Δ$ for Direct Weight Mixing: Post-Train Large Language Model at Zero Cost
Abstract page for arXiv paper 2504.21023: Param$Δ$ for Direct Weight Mixing: Post-Train Large Language Model at Zero Cost
arXiv - AI
·
4 min
·
Llms
[2603.03258] Inherited Goal Drift: Contextual Pressure Can Undermine Agentic Goals
Abstract page for arXiv paper 2603.03258: Inherited Goal Drift: Contextual Pressure Can Undermine Agentic Goals
arXiv - AI
·
4 min
·
Llms
[2603.02635] SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety
Abstract page for arXiv paper 2603.02635: SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety
arXiv - Machine Learning
·
4 min
·
Llms
[2603.03242] Density-Guided Response Optimization: Community-Grounded Alignment via Implicit Acceptance Signals
Abstract page for arXiv paper 2603.03242: Density-Guided Response Optimization: Community-Grounded Alignment via Implicit Acceptance Signals
arXiv - AI
·
4 min
·
Llms
[2603.02630] MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
Abstract page for arXiv paper 2603.02630: MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
arXiv - AI
·
4 min
·
Llms
[2603.03233] AI-for-Science Low-code Platform with Bayesian Adversarial Multi-Agent Framework
Abstract page for arXiv paper 2603.03233: AI-for-Science Low-code Platform with Bayesian Adversarial Multi-Agent Framework
arXiv - AI
·
4 min
·
Llms
[2603.03203] No Memorization, No Detection: Output Distribution-Based Contamination Detection in Small Language Models
Abstract page for arXiv paper 2603.03203: No Memorization, No Detection: Output Distribution-Based Contamination Detection in Small Langu...
arXiv - AI
·
3 min
·
Llms
[2603.02604] Heterogeneous Agent Collaborative Reinforcement Learning
Abstract page for arXiv paper 2603.02604: Heterogeneous Agent Collaborative Reinforcement Learning
arXiv - Machine Learning
·
3 min
·
Llms
[2603.03175] Saarthi for AGI: Towards Domain-Specific General Intelligence for Formal Verification
Abstract page for arXiv paper 2603.03175: Saarthi for AGI: Towards Domain-Specific General Intelligence for Formal Verification
arXiv - AI
·
4 min
·
Llms
[2603.03147] Agentic AI-based Coverage Closure for Formal Verification
Abstract page for arXiv paper 2603.03147: Agentic AI-based Coverage Closure for Formal Verification
arXiv - AI
·
3 min
·
Llms
[2603.03080] Beyond Factual Correctness: Mitigating Preference-Inconsistent Explanations in Explainable Recommendation
Abstract page for arXiv paper 2603.03080: Beyond Factual Correctness: Mitigating Preference-Inconsistent Explanations in Explainable Reco...
arXiv - AI
·
3 min
·
Llms
[2603.03116] Beyond Task Completion: Revealing Corrupt Success in LLM Agents through Procedure-Aware Evaluation
Abstract page for arXiv paper 2603.03116: Beyond Task Completion: Revealing Corrupt Success in LLM Agents through Procedure-Aware Evaluation
arXiv - AI
·
4 min
·
Llms
[2603.02510] ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolution
Abstract page for arXiv paper 2603.02510: ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evol...
arXiv - Machine Learning
·
4 min
·
Llms
[2603.03002] SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models
Abstract page for arXiv paper 2603.03002: SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models
arXiv - AI
·
4 min
·
Llms
[2603.02482] MUSE: A Run-Centric Platform for Multimodal Unified Safety Evaluation of Large Language Models
Abstract page for arXiv paper 2603.02482: MUSE: A Run-Centric Platform for Multimodal Unified Safety Evaluation of Large Language Models
arXiv - Machine Learning
·
4 min
·
Llms
[2603.03078] RAPO: Expanding Exploration for LLM Agents via Retrieval-Augmented Policy Optimization
Abstract page for arXiv paper 2603.03078: RAPO: Expanding Exploration for LLM Agents via Retrieval-Augmented Policy Optimization
arXiv - AI
·
4 min
·
Llms
[2603.03072] TikZilla: Scaling Text-to-TikZ with High-Quality Data and Reinforcement Learning
Abstract page for arXiv paper 2603.03072: TikZilla: Scaling Text-to-TikZ with High-Quality Data and Reinforcement Learning
arXiv - AI
·
4 min
·
Llms
[2603.03018] REGAL: A Registry-Driven Architecture for Deterministic Grounding of Agentic AI in Enterprise Telemetry
Abstract page for arXiv paper 2603.03018: REGAL: A Registry-Driven Architecture for Deterministic Grounding of Agentic AI in Enterprise T...
arXiv - AI
·
4 min
·
Llms
[2603.03005] OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Structured Agents
Abstract page for arXiv paper 2603.03005: OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Struct...
arXiv - AI
·
4 min
·
Related Topics
Stay updated with AI News
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime