Posts

All the articles I've posted.

6.0
Neural Bandit Based Optimal LLM Selection for a Pipeline of Subtasks
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
Foundational Design Principles and Patterns for Building Robust and Adaptive GenAI-Native Systems
2026年04月23日
· arXiv· 04/23 08:00 采集
7.0
Task-Stratified Knowledge Scaling Laws for Post-Training Quantized Large Language Models
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
Do Hallucination Neurons Generalize? Evidence from Cross-Domain Transfer in LLMs
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
OThink-SRR1: Search, Refine and Reasoning with Reinforced Learning for Large Language Models
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
Cognis: Context-Aware Memory for Conversational AI Agents
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
Self-Describing Structured Data with Dual-Layer Guidance: A Lightweight Alternative to RAG for Precision Retrieval in Large-Scale LLM Knowledge Navigation
2026年04月23日
· arXiv· 04/23 08:00 采集
7.0
Avoiding Overthinking and Underthinking: Curriculum-Aware Budget Scheduling for LLMs
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
From Recall to Forgetting: Benchmarking Long-Term Memory for Personalized Agents
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
TriEx: A Game-based Tri-View Framework for Explaining Internal Reasoning in Multi-Agent LLMs
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
Less Languages, Less Tokens: An Efficient Unified Logic Cross-lingual Chain-of-Thought Reasoning Framework
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
To Know is to Construct: Schema-Constrained Generation for Agent Memory
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
Chasing the Public Score: User Pressure and Evaluation Exploitation in Coding Agent Workflows
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
Hybrid Policy Distillation for LLMs
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
Stateless Decision Memory for Enterprise AI Agents
2026年04月23日
· arXiv· 04/23 08:00 采集
6.0
FSFM: A Biologically-Inspired Framework for Selective Forgetting of Agent Memory
2026年04月23日
· arXiv· 04/23 08:00 采集
6.5
SWE-chat: Coding Agent Interactions From Real Users in the Wild
2026年04月23日
· arXiv· 04/23 08:00 采集

Posts

Neural Bandit Based Optimal LLM Selection for a Pipeline of Subtasks

Foundational Design Principles and Patterns for Building Robust and Adaptive GenAI-Native Systems

Task-Stratified Knowledge Scaling Laws for Post-Training Quantized Large Language Models

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation

Do Hallucination Neurons Generalize? Evidence from Cross-Domain Transfer in LLMs

OThink-SRR1: Search, Refine and Reasoning with Reinforced Learning for Large Language Models

Cognis: Context-Aware Memory for Conversational AI Agents

Self-Describing Structured Data with Dual-Layer Guidance: A Lightweight Alternative to RAG for Precision Retrieval in Large-Scale LLM Knowledge Navigation

Avoiding Overthinking and Underthinking: Curriculum-Aware Budget Scheduling for LLMs

From Recall to Forgetting: Benchmarking Long-Term Memory for Personalized Agents

TriEx: A Game-based Tri-View Framework for Explaining Internal Reasoning in Multi-Agent LLMs

Less Languages, Less Tokens: An Efficient Unified Logic Cross-lingual Chain-of-Thought Reasoning Framework

To Know is to Construct: Schema-Constrained Generation for Agent Memory

All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG

Chasing the Public Score: User Pressure and Evaluation Exploitation in Coding Agent Workflows

Hybrid Policy Distillation for LLMs

JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents

Stateless Decision Memory for Enterprise AI Agents

FSFM: A Biologically-Inspired Framework for Selective Forgetting of Agent Memory

SWE-chat: Coding Agent Interactions From Real Users in the Wild