Liang He

From Relevance to Execution Utility: Reward-Aware Dynamic Execution Gating for Skill-Based LLM Agents

Liang He, Jingbo Wen, Hongyu Gu, Hao Li et autres

Agent skills are increasingly used to equip large language model (LLM) agents with reusable procedural knowledge. Although recent work has substantially improved skill retrieval due to the increasing skill libraries, retrieving a plausible skill bundle does not guarantee that executing it is …

Not All Visual Tokens Are Equally Safe to Remove:Consequence-Sensitive Visual Token Compression

Jingbo Wen, Liang He, Mingyu Cao, Haoyu Wang et autres

Visual token compression for vision--language models (VLMs) has largely relied on criteria such as attention, redundancy, and uncertainty to maximize average accuracy under a fixed compute budget, implicitly assuming that all errors carry equal cost. However, the consequence of an incorrect prediction …

DataShield: Uncovering Risky Fine-Tuning Data Across LLMs Through Consensus Subspace Alignment

Zefeng Wu, Weiwei Qi, Jielong Chen, Tianhang Zheng et autres

Fine-tuning large language models (LLMs) on domain-specific datasets has become a standard paradigm for adapting LLMs to specialized applications. However, recent work has shown that even fine-tuning on benign task-specific data can substantially weaken the safety capabilities of LLMs. While existing efforts …

cn (code pays fourni par la source)

NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs

Jiarong Zhao, Zhikai Lei, Zhiheng Xi, Rui Zheng et autres

Scaling executable agent training data for LLM post-training is bottlenecked by substrate-bound methods that tie task generation to predefined tools, repositories, or skill graphs: expanding coverage requires manual substrate engineering, each new domain demands a bespoke pipeline, and the resulting task distributions …

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui et autres

We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure …

Anything2Skill: Compiling External Knowledge into Reusable Skills for Agents

Qianjun Pan, Yutao Yang, Junsong Li, Jie Zhou et autres

Retrieval-augmented generation (RAG) enables agents to access external knowledge at inference time, but it primarily retrieves fragmented declarative evidence, leaving agents to repeatedly infer task procedures from passages, manuals, examples, logs, or trajectories. This raises a fundamental question: can skills extracted from …

MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery

Shangheng Du, Xiangchao Yan, Jinxin Shi, Zongsheng Cao et autres

Large language model (LLM) agents are increasingly applied to long-horizon tasks such as scientific discovery and machine learning engineering (MLE), where sustained self-evolution becomes a key capability. However, existing MLE agents suffer from inter-branch information isolation, memoryless search, and lack of hierarchical …

A Survey on the Optimization of Large Language Model-based Agents

Shangheng Du, Jiabao Zhao, Jinxin Shi, Zhentao Xie et autres

With the rapid development of Large Language Models (LLMs), LLM-based agents have been widely adopted in various fields, becoming essential for autonomous decision-making and interactive tasks. However, current work typically relies on prompt design or fine-tuning strategies applied to vanilla LLMs, which …

cn (code pays fourni par la source)