Manling Li

Theory of Space: Can Foundation Models Construct Spatial Beliefs through Active Exploration?

Weak-to-Strong Generalization with Failure Trajectories

ROSETTA: Constructing Code-Based Reward from Unconstrained Language Preference

ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment

Spatial Mental Modeling from Limited Views

ENACT: Evaluating Embodied Cognition with World Modeling of Egocentric Interaction

Your Language Model Secretly Contains Personality Subnetworks

Internalizing World Models via Self-Play Finetuning for Agentic RL

SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of LLM-based Embodied Agents

Observational Scaling Laws in LLM-based Embodied Decision Making

Unary Feedback as Observation: Incentivizing Self-Reflection in Large Language Models via Multi-Turn RL

CURV: Enhancing Chart Understanding through Visual Grounded Reasoning

ERA: Transforming VLMs into Embodied Agents via Embodied Prior Learning and Online Reinforcement Learning

Probing the Limits of Embodied Spatial Planning in LLMs

Federated Agent Reinforcement Learning

AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning

Conv-CoA: Open-domain Question Answering via Conversational Chain-of-Action with Hopfield Retriever

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents

Exploring Diffusion Transformer Designs via Grafting

Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models

SyncMind: Measuring Agent Out-of-Sync Recovery in Collaborative Software Engineering

Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas

Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging

Visually Descriptive Language Model for Vector Graphics Reasoning

MentalArena: Self-play Training of Language Models for Diagnosis and Treatment of Mental Health Disorders

AdaptVis: Spatial Understanding in Vision-Language Models Requires Adaptive Attention

Multlingual Abstractive Event Extraction for the Real World