Ben's AI Feed

arXiv cs.CL models Research Paper score 118.2 2026-03-23

From Tokens To Agents: A Researcher's Guide To Understanding Large Language Models

arXiv cs.CL models Research Paper score 110.2 2026-03-23

Structured Prompting for Arabic Essay Proficiency: A Trait-Centric Evaluation Approach

arXiv cs.AI agents Research Paper score 108.2 2026-03-23

The α-Law of Observable Belief Revision in Large Language Model Inference

arXiv cs.CL multimodal Research Paper score 105.2 2026-03-23

L2V-CoT: Cross-Modal Transfer of Chain-of-Thought Reasoning via Latent Intervention

arXiv cs.CL models Research Paper score 99.2 2026-03-23

FedPDPO: Federated Personalized Direct Preference Optimization for Large Language Model Alignment

arXiv cs.CL models Research Paper score 97.2 2026-03-23

DLLM Agent: See Farther, Run Faster

arXiv cs.CL models Research Paper score 96.2 2026-03-23

Taking a Deep Breath: Enhancing Language Modeling of Large Language Models with Sentinel Tokens

arXiv cs.LG models Research Paper score 94.2 2026-03-23

FastMMoE: Accelerating Multimodal Large Language Models through Dynamic Expert Activation and Routing-Aware Token Pruning

arXiv cs.CL models Research Paper score 94.2 2026-03-23

From Comprehension to Reasoning: A Hierarchical Benchmark for Automated Financial Research Reporting

arXiv cs.CL models Research Paper score 94.2 2026-03-23

Test-Time Alignment for Large Language Models via Textual Model Predictive Control

arXiv cs.AI models Research Paper score 91.2 2026-03-23

PowerLens: Taming LLM Agents for Safe and Personalized Mobile Power Management

arXiv cs.LG models Research Paper score 89.2 2026-03-23

The Autonomy Tax: Defense Training Breaks LLM Agents

arXiv cs.CL agents Research Paper score 89.2 2026-03-23

ReViSQL: Achieving Human-Level Text-to-SQL

arXiv cs.CL models Research Paper score 89.2 2026-03-23

Identifying and Mitigating Bottlenecks in Role-Playing Agents: A Systematic Study of Disentangling Character Profile Axes

arXiv cs.LG models Research Paper score 88.2 2026-03-23

MAPLE: Metadata Augmented Private Language Evolution

arXiv cs.CL models Research Paper score 88.2 2026-03-23

Enhancing Legal LLMs through Metadata-Enriched RAG Pipelines and Direct Preference Optimization