12 papers
Self-Evolving Agent Harness
Chuyang Wei, Haoxiang Guan, Minghang Zhu +5
Harness self-evolution revises an agent's software scaffolding, including prompts, memory, workflows, and executable components, while keeping model weights fixed. We formalize a f…
PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models
DeepCybo Team, Yu Bin, Haipeng Cao +51
We present PhysBrain 1.5, a unified model for understanding physical environments, generating actions, and predicting future states. Motivated by the physical loop of observation,…
ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search
Jiyan He, Guang Liang, Hao Liu +19
In this work, we present ZGCM-1, a fully open 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency. ZGCM-1 is founded on a core prem…
DEEPRUBRIC: Evidence-Tree Rubric Supervision for Efficient Reinforcement Learning of Deep Research Agents
Minghang Zhu, Chuyang Wei, Junhao Xu +3
Deep research agents synthesize long-form reports by searching and reasoning over retrieved evidence. Reinforcement learning with rubric-based rewards improves these agents by opti…
FutureWorld: A Live Reinforcement Learning Environment for Predictive Agents with Real-World Outcome Rewards
Zhixin Han, Yanzhi Zhang, Chuyang Wei +11
Live future prediction refers to the task of making predictions about real-world events before they unfold. This task is increasingly studied using large language model-based agent…
Harnessing Pre-Resolution Signals for Future Prediction Agents
Chuyang Wei, Maohang Gao, Zhixin Han +12
Many high-stakes decisions depend on forecasts made before outcomes are known. In this future prediction setting, the central challenge is that public evidence evolves over time, w…
Can a Lightweight Automated AI Pipeline Solve Research-Level Mathematical Problems?
Lve Meng, Weilong Zhao, Yanzhi Zhang +2
Large language models (LLMs) have recently achieved remarkable success in generating rigorous mathematical proofs, with "AI for Math" emerging as a vibrant field of research (Ju et…
Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math Reasoning
Yanzhi Zhang, Yitong Duan, Zhaoxi Zhang +2
Test-time scaling has emerged as a promising direction for enhancing the reasoning capabilities of Large Language Models in last few years. In this work, we propose Population-Evol…
One Tool Is Enough: Reinforcement Learning for Repository-Level LLM Agents
Zhaoxi Zhang, Yitong Duan, Yanzhi Zhang +9
Locating files and functions requiring modification in large software repositories is challenging due to their scale and structural complexity. Existing LLM-based methods typically…
Sparse Autoencoders Reveal Interpretable Structure in Small Gene Language Models
Haoxiang Guan, Jiyan He, Jie Zhang
Sparse autoencoders (SAEs) have recently emerged as a powerful tool for interpreting the internal representations of large language models (LLMs), revealing latent latent features…
No Free Lunch: Rethinking Internal Feedback for LLM Reasoning
Yanzhi Zhang, Zhaoxi Zhang, Haoxiang Guan +6
Reinforcement learning has emerged as a powerful paradigm for post-training large language models (LLMs) to improve reasoning. Approaches like Reinforcement Learning from Human Fee…
RedChronos: A Large Language Model-Based Log Analysis System for Insider Threat Detection in Enterprises
Chenyu Li, Zhengjia Zhu, Jiyan He +1
Internal threat detection (IDT) aims to address security threats within organizations or enterprises by identifying potential or already occurring malicious threats within vast amo…