9 papers
CONTRAMEM: Learning Self-Evolving Procedural Memory from Contrasting Multi-Model Trajectories
Zheyuan Deng, Binghang Lu, Hanqi Feng +11
Autonomous computer-use agents are increasingly applied to long-horizon tasks requiring coordinated application calls, persistent state tracking, and verifier-sensitive writes, yet…
MicroVerse: An Instrument for Measuring Self-Authored Identity Drift in Long-Horizon Multi-Agent Language-Model Simulations
Sky Ng, Brihi Joshi, Ishan Gupta +47
Long-horizon, multi-agent language model (LM) simulations are widely proposed for studying social behavior, yet instruments to measure whether persona-conditioned agents maintain i…
Co-Evolving Skill Generation and Policy Optimization
Zhiwei Zhang, Yudi Lin, Nikki Lijing Kuang +4
Skill-augmented reinforcement learning improves language agents by storing reusable procedural knowledge acquired from past experience. Existing methods typically use strong langua…
Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering
Xiaomin Li, Jianheng Hou, Zheyuan Deng +6
Large reasoning models (LRMs) increasingly expose chain-of-thought-like reasoning for transparency, verification, and deliberate problem solving. This creates a safety blind spot:…
InEdit-Bench: Benchmarking Intermediate Logical Pathways for Intelligent Image Editing Models
Zhiqiang Sheng, Xumeng Han, Zhiwei Zhang +6
Multimodal generative models have made significant strides in image editing, demonstrating impressive performance on a variety of static tasks. However, their proficiency typically…
Multi-Head Low-Rank Attention
Songtao Liu, Hongwu Peng, Zhiwei Zhang +2
Long-context inference in large language models is bottlenecked by Key--Value (KV) cache loading during the decoding stage, where the sequential nature of generation requires repea…
Bradley-Terry and Multi-Objective Reward Modeling Are Complementary
Zhiwei Zhang, Hui Liu, Xiaomin Li +10
Reward models trained on human preference data have demonstrated strong effectiveness in aligning Large Language Models (LLMs) with human intent under the framework of Reinforcemen…
Data-adaptive Safety Rules for Training Reward Models
Xiaomin Li, Mingye Gao, Zhiwei Zhang +2
Reinforcement Learning from Human Feedback (RLHF) is commonly employed to tailor models to human preferences, especially to improve the safety of outputs from large language models…
Selection of LLM Fine-Tuning Data based on Orthogonal Rules
Xiaomin Li, Mingye Gao, Zhiwei Zhang +2
High-quality training data is critical to the performance of large language models (LLMs). Recent work has explored using LLMs to rate and select data based on a small set of human…