Publications (19)
NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?
Yuru Wang, Lejun Cheng, Yuxin Zuo +14
We introduce NatureBench, a cross-discipline benchmark of 90 tasks distilled from peer-reviewed Nature-family publications, designed to evaluate whether AI coding agents can move b…
EnterpriseClawBench: Benchmarking Agents from Real Workplace Sessions
Jincheng Zhong, Weizhi Wang, Che Jiang +5
Enterprise agents increasingly operate inside workspaces: they read heterogeneous files, invoke tools, and deliver business artifacts. We introduce EnterpriseClawBench, an enterpri…
FlowRL: Matching Reward Distributions for LLM Reasoning
Xuekai Zhu, Daixuan Cheng, Dinghuai Zhang +20
We propose FlowRL: matching the full reward distribution via flow balancing instead of maximizing rewards in large language model (LLM) reinforcement learning (RL). Recent advanced…
DePass: Unified Feature Attributing by Simple Decomposed Forward Pass
Xiangyu Hong, Che Jiang, Kai Tian +4
Attributing the behavior of Transformer models to internal computations is a central challenge in mechanistic interpretability. We introduce DePass, a unified framework for feature…
WorldGenBench: A World-Knowledge-Integrated Benchmark for Reasoning-Driven Text-to-Image Generation
Daoan Zhang, Che Jiang, Ruoshi Xu +7
Recent advances in text-to-image (T2I) generation have achieved impressive results, yet existing models still struggle with prompts that require rich world knowledge and implicit r…
Frontis-MA1: Training an AI4AI Model towards Recursive Self-Improvement in Machine Learning Engineering
Junlin Yang, Che Jiang, Yu Fu +21
The paper presents Frontis-MA1, a 35‑billion‑parameter model trained as a meta‑evolution agent for machine learning engineering, using a new OpenMLE stack that combines operator le…
A Survey of Reinforcement Learning for Large Reasoning Models
Kaiyan Zhang, Yuxin Zuo, Bingxiang He +36
In this paper, we survey recent advances in Reinforcement Learning (RL) for reasoning with Large Language Models (LLMs). RL has achieved remarkable success in advancing the frontie…
TIR-Agent: Training an Explorative and Efficient Agent for Image Restoration
Guoli Jia, Yisheng Zhang, Haote Hu +11
Vision-language agents that orchestrate specialized tools for image restoration (IR) have emerged as a promising method, yet most existing frameworks operate in a training-free man…
ReportQA: QA-Based Radiology Report Evaluation
Yiming Shi, Shaoshuai Yang, Xi Chen +10
Radiology report evaluation is essential for advancing automated report generation. Natural language generation metrics have limited clinical relevance. Clinical efficacy (CE) metr…
AdsQA: Towards Advertisement Video Understanding
Xinwei Long, Kai Tian, Peng Xu +10
Large language models (LLMs) have taken a great step towards AGI. Meanwhile, an increasing number of domain-specific problems such as math and programming boost these general-purpo…
MSI-Agent: Incorporating Multi-Scale Insight into Embodied Agents for Superior Planning and Decision-Making
Dayuan Fu, Biqing Qi, Yihuai Gao +3
Long-term memory is significant for agents, in which insights play a crucial role. However, the emergence of irrelevant insight and the lack of general insight can greatly undermin…
On Large Language Models' Hallucination with Regard to Known Facts
Che Jiang, Biqing Qi, Xiangyu Hong +6
Large language models are successful in answering factoid questions but are also prone to hallucination. We investigate the phenomenon of LLMs possessing correct answer knowledge y…
SSRL: Self-Search Reinforcement Learning
Yuchen Fan, Kaiyan Zhang, Heng Zhou +15
We investigate the potential of large language models (LLMs) to serve as efficient simulators for agentic search tasks in reinforcement learning (RL), thereby reducing dependence o…
HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction
Yunze Liu, Yun Liu, Che Jiang +7
We present HOI4D, a large-scale 4D egocentric dataset with rich annotations, to catalyze the research of category-level human-object interaction. HOI4D consists of 2.4M RGB-D egoce…
Quantum sensing of rotation velocity based on Bose-Hubbard model
Che Jiang, Yaojie Zeng, Qi Qin +2
This work theoretically study the Bose-Hubbard model in a ring geometry in a rotating frame. We obtain an effective Hamiltonian by using unitary transformation, where the effect of…
Fourier Position Embedding: Enhancing Attention's Periodic Extension for Length Generalization
Ermo Hua, Che Jiang, Xingtai Lv +7
Extending the context length of Language Models (LMs) by improving Rotary Position Embedding (RoPE) has become a trend. While prior works mainly address RoPE's limitations within a…
Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism
Yuhua Jiang, Shuang Cheng, Yihao Liu +7
Large Language Models (LLMs) excel at general language tasks but struggle in specialized domains. Specialized Generalist Models (SGMs) address this by preserving broad capabilities…
On the token distance modeling ability of higher RoPE attention dimension
Xiangyu Hong, Che Jiang, Biqing Qi +4
Length extrapolation algorithms based on Rotary position embedding (RoPE) have shown promising results in extending the context length of language models. However, understanding ho…
Context-Aware Autoregressive Models for Multi-Conditional Image Generation
Yixiao Chen, Zhiyuan Ma, Guoli Jia +3
Autoregressive transformers have recently shown impressive image generation quality and efficiency on par with state-of-the-art diffusion models. Unlike diffusion architectures, au…