20 papers
AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models
Rui Wang, Xiangyu Wang, Donglin Yang +4
World-action models (WAMs) couple predictive visual modeling with action generation, typically relying on iterative denoising with a fixed denoising steps. However, manipulation ta…
ActiveArena: Benchmarking and Understanding Active Perception in Robotic Manipulation
Yibo Li, Enshen Zhou, Rui Chen +7
Active perception and manipulation are crucial for robots to interact with complex scenes. Existing benchmarks struggle to evaluate how robots effectively acquire and maintain info…
PhysAgent: Reflective Agentic Physics Control for Physically Plausible Video Generation
Qirui Li, Jinkun Hao, Yibo Li +3
Recent advances in physics-grounded video generation leverage physics simulation as a physical prior to guide video synthesis toward physically plausible outcomes. The simulation p…
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
Shuo Ji, Yibo Li, Bryan Hooi
Despite recent progress, LLM agents still struggle with reasoning over long interaction histories. While current memory-augmented agents rely on a static retrieve-then-reason parad…
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…
WARD: Adversarially Robust Defense of Web Agents Against Prompt Injections
Tri Cao, Yulin Chen, Hieu Cao +8
Web agents can autonomously complete online tasks by interacting with websites, but their exposure to open web environments makes them vulnerable to prompt injection attacks embedd…
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…
WebAgentGuard: A Reasoning-Driven Guard Model for Detecting Prompt Injection Attacks in Web Agents
Yulin Chen, Tri Cao, Haoran Li +7
Web agents powered by vision-language models (VLMs) enable autonomous interaction with web environments by perceiving and acting on both visual and textual webpage content to accom…
Meta-TTL: Meta-Learning Self-Improvement Policies for Language Agents
Zhanzhi Lou, Hui Chen, Yibo Li +2
Test-Time Learning (TTL) enables language agents to iteratively refine their performance through repeated interactions with the environment at inference time. At the core of TTL is…
MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome
Fangda Ye, Yuxin Hu, Pengxiang Zhu +19
Recent progress in deep research systems has been impressive, but evaluation still lags behind real user needs. Existing benchmarks predominantly assess final reports using fixed r…
Towards Natural Personalization: Evaluating Long-Horizon Preference Following in Personalized User-LLM Interactions
Qianyun Guo, Yibo Li, Yue Liu +1
Large Language Models (LLMs) are increasingly serving as personal assistants, where users may share individual preferences over extended interactions. However, assessing how well L…
GuardReasoner-Omni: A Reasoning-based Multi-modal Guardrail for Text, Image, Video, and Audio
Zhenhao Zhu, Yue Liu, Yanpei Guo +9
We present GuardReasoner-Omni, a reasoning-based guardrail model designed to moderate text, image, video, and audio data. First, we construct a comprehensive training corpus compri…
EvoClinician: A Self-Evolving Agent for Multi-Turn Medical Diagnosis via Test-Time Evolutionary Learning
Yufei He, Juncheng Liu, Zhiyuan Hu +9
Prevailing medical AI operates on an unrealistic ''one-shot'' model, diagnosing from a complete patient file. However, real-world diagnosis is an iterative inquiry where Clinicians…
RoboBrain 2.5: Depth in Sight, Time in Mind
Huajie Tan, Enshen Zhou, Zhiyu Li +32
We introduce RoboBrain 2.5, a next-generation embodied AI foundation model that advances general perception, spatial reasoning, and temporal modeling through extensive training on…
Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates
Yibo Li, Zijie Lin, Ailin Deng +5
While Large Language Model (LLM) agents excel at general tasks, they inherently struggle with continual adaptation due to the frozen weights after deployment. Conventional reinforc…
Towards Spatial Trace with Reasoning in Vision-Language Models for Robotics
Enshen Zhou, Yibo Li, Jingkun An +12
Spatial tracing, as a fundamental embodied interaction ability for robots, is inherently challenging as it requires multi-step metric-grounded reasoning compounded with complex spa…
EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems
Yufei He, Juncheng Liu, Yue Liu +5
A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like "clever but clueless interns" in novel environ…
MagicVL-2B: Empowering Vision-Language Models on Mobile Devices with Lightweight Visual Encoders via Curriculum Learning
Yi Liu, Xiao Xu, Zeyu Xu +10
Vision-Language Models (VLMs) have achieved remarkable breakthroughs in recent years, enabling a diverse array of applications in everyday life. However, the substantial computatio…
ConfTuner: Training Large Language Models to Express Their Confidence Verbally
Yibo Li, Miao Xiong, Jiaying Wu +1
Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare, where accurate expressions of uncertainty are essential for reli…
MLR-Bench: Evaluating AI Agents on Open-Ended Machine Learning Research
Hui Chen, Miao Xiong, Yujie Lu +7
Recent advancements in AI agents have demonstrated their growing potential to drive and support scientific discovery. In this work, we introduce MLR-Bench, a comprehensive benchmar…