12 papers
When Does a Skill Add Value? Task-Conditional Gain Prediction for Selective Skill Use
Anjie Xu, Zhiyu Zhang, Ruiqing Ding +2
Agent skills are expected to improve task performance. Yet we find that they often provide no benefit, and can even hurt performance while incurring additional token costs. Can we…
Ideation Arena: Evaluating LLM Generated Research Ideas with Battle-style Human Expert Assessment
Zhiyu Chen, Keyu Zhao, Jigao Fu +8
Evaluating research ideas generated by LLMs is difficult because their scientific value cannot be fully determined by objective criteria, and no single reference answer specifies w…
Markets, Not Planners: Decentralized Orchestration of LLM Agents with Private Information
Xiao Liu, Haoyang Li, Songwei Li +4
As LLM agents proliferate, built by different parties and with different capabilities and costs, orchestrating them is more like assembling labor across the economy than a computer…
LitReview Arena: Evaluating Literature Review Agents with Battle-Style Peer Review Platform
Ruotong Zhao, Zhiyu Chen, Xurui Liu +7
Literature reviews are essential to scientific progress, but rigorously evaluating automatically generated reviews remains difficult because many aspects of research utility depend…
AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discovery
Yu Li, Chenyang Shao, Xinyang Liu +13
Artificial intelligence research increasingly depends on prolonged cycles of reproduction, debugging, and iterative refinement to achieve State-Of-The-Art (SOTA) performance, creat…
LiveFigure: Generating Editable Scientific Illustration with VLM Agents
Chenyang Shao, Jiahe Liu, Fengli Xu +1
Scientific illustrations are essential for depicting conceptual designs, methodologies, and experimental workflows in research, playing a pivotal role in communicating complex acad…
ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning
Wanghan Xu, Yuhao Zhou, Hengyuan Zhao +8
Large language models can fail in critic interaction not only by answering incorrectly, but also by abandoning an initially correct scientific solution after user criticism. This i…
PhysOmni: Physics-Grounded Multi-Object Scene Generation from a Single Image with Real-Time Interaction
Xin Zhang, Yabo Chen, Yijie Fang +5
Recent generative video models achieve impressive visual quality but remain constrained by limited physical consistency and controllability. Existing video generation methods provi…
Diffuse Thinking: Exploring Diffusion Language Models as Efficient Thought Proposers for Reasoning
Chenyang Shao, Sijian Ren, Fengli Xu +1
In recent years, large language models (LLMs) have witnessed remarkable advancements, with the test-time scaling law consistently enhancing the reasoning capabilities. Through syst…
AgentStealth: Reinforcing Large Language Model for Anonymizing User-generated Text
Chenyang Shao, Tianxing Li, Chenhao Pu +2
In today's digital world, casual user-generated content often contains subtle cues that may inadvertently expose sensitive personal attributes. Such risks underscore the growing im…
Division-of-Thoughts: Harnessing Hybrid Language Model Synergy for Efficient On-Device Agents
Chenyang Shao, Xinyuan Hu, Yutang Lin +1
The rapid expansion of web content has made on-device AI assistants indispensable for helping users manage the increasing complexity of online tasks. The emergent reasoning ability…
LIMP: Large Language Model Enhanced Intent-aware Mobility Prediction
Songwei Li, Jie Feng, Jiawei Chi +3
Human mobility prediction is essential for applications like urban planning and transportation management, yet it remains challenging due to the complex, often implicit, intentions…