学术前沿

1253 篇文章

6.0
Semantic Structure of Feature Space in Large Language Models
2026年05月01日
· 05/01 14:31 采集
7.0
Implicit bias produces neural scaling laws in learning curves, from perceptrons to deep networks
2026年05月01日
· 05/01 14:31 采集
6.0
Exploring the Limits of Pruning: Task-Specific Neurons, Model Collapse, and Recovery in Task-Specific Large Language Models
2026年05月01日
· 05/01 14:31 采集
7.0
Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs 「Difficult「 Downstream Tasks in LLMs
2026年05月01日
· 05/01 14:31 采集
8.0
In-Context Prompting Obsoletes Agent Orchestration for Procedural Tasks
2026年05月01日
· 05/01 14:31 采集
7.0
Agentic Education: Using Claude Code to Teach Claude Code
2026年05月01日
· 05/01 14:31 采集
7.0
Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents
2026年05月01日
· 05/01 14:31 采集
6.0
Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents
2026年05月01日
· 05/01 14:31 采集
6.0
Learning to Aggregate Zero-Shot LLM Agents for Corporate Disclosure Classification
2026年05月01日
· 05/01 14:31 采集
6.0
Semantic Structure of Feature Space in Large Language Models
2026年05月01日
· 05/01 14:31 采集
6.0
Agentic AI for Cybersecurity: A Meta-Cognitive Architecture for Governable Autonomy
2026年05月01日
· 05/01 14:31 采集
6.0
When Your LLM Reaches End-of-Life: A Framework for Confident Model Migration in Production Systems
2026年05月01日
· 05/01 14:31 采集
6.0
Lost in Space? Vision-Language Models Struggle with Relative Camera Pose Estimation
2026年05月01日
· 05/01 14:31 采集
6.0
Taxon: Hierarchical Tax Code Prediction with Semantically Aligned LLM Expert Guidance
2026年05月01日
· 05/01 14:31 采集
6.0
Context Matters: Peer-Aware Student Behavioral Engagement Measurement via VLM Action Parsing and LLM Sequence Classification
2026年05月01日
· 05/01 14:31 采集
6.0
Exploration Hacking: Can LLMs Learn to Resist RL Training?
2026年05月01日
· 05/01 14:31 采集
6.0
Explainable Load Forecasting with Covariate-Informed Time Series Foundation Models
2026年05月01日
· 05/01 14:31 采集
7.0
Global Optimality for Constrained Exploration via Penalty Regularization
2026年05月01日
· 05/01 14:31 采集
6.0
Kernelized Advantage Estimation: From Nonparametric Statistics to LLM Reasoning
2026年05月01日
· 05/01 14:31 采集
6.0
Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretraining
2026年05月01日
· 05/01 14:31 采集

学术前沿

Semantic Structure of Feature Space in Large Language Models

Implicit bias produces neural scaling laws in learning curves, from perceptrons to deep networks

Exploring the Limits of Pruning: Task-Specific Neurons, Model Collapse, and Recovery in Task-Specific Large Language Models

Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs 「Difficult「 Downstream Tasks in LLMs

In-Context Prompting Obsoletes Agent Orchestration for Procedural Tasks

Agentic Education: Using Claude Code to Teach Claude Code

Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents

Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents

Learning to Aggregate Zero-Shot LLM Agents for Corporate Disclosure Classification

Semantic Structure of Feature Space in Large Language Models

Agentic AI for Cybersecurity: A Meta-Cognitive Architecture for Governable Autonomy

When Your LLM Reaches End-of-Life: A Framework for Confident Model Migration in Production Systems

Lost in Space? Vision-Language Models Struggle with Relative Camera Pose Estimation

Taxon: Hierarchical Tax Code Prediction with Semantically Aligned LLM Expert Guidance

Context Matters: Peer-Aware Student Behavioral Engagement Measurement via VLM Action Parsing and LLM Sequence Classification

Exploration Hacking: Can LLMs Learn to Resist RL Training?

Explainable Load Forecasting with Covariate-Informed Time Series Foundation Models

Global Optimality for Constrained Exploration via Penalty Regularization

Kernelized Advantage Estimation: From Nonparametric Statistics to LLM Reasoning

Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretraining