14 papers
APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents
Yibo Li, Jiashuo Yang, Zhi Zheng +5
LLM agents have shown strong performance across a wide range of complex tasks, including interactive environments that require long-horizon decision making. But these agents cannot…
TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering
Yuan Sui, Yulin Chen, Yibo Li +6
When language model agents tackle complex software engineering tasks, they often degrade over long trajectories, which we define as *agent drift*. We focus on two recurring failure…
Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation
Yuan Sui, Bryan Hooi
Training large language models (LLMs) for non-verifiable tasks, such as creative writing, dialogue, and ethical reasoning, remains challenging due to the absence of ground-truth la…
Meta-Reasoner: Dynamic Guidance for Optimized Inference-time Reasoning in Large Language Models
Yuan Sui, Yufei He, Tri Cao +3
Large Language Models (LLMs) often struggle with computational efficiency and error propagation in multi-step reasoning tasks. While recent advancements on prompting and post-train…
Robustness via Referencing: Defending against Prompt Injection Attacks by Referencing the Executed Instruction
Yulin Chen, Haoran Li, Yuan Sui +8
Large language models (LLMs) have demonstrated impressive performance and have come to dominate the field of natural language processing (NLP) across various tasks. However, due to…
VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents
Tri Cao, Bennett Lim, Yue Liu +7
Computer-Use Agents (CUAs) with full system access enable powerful task automation but pose significant security and privacy risks due to their ability to manipulate files, access…