Sheng Zhong

Sheng Zhong, Nima Fazeli, Dmitry Berenson

This paper presents Rummaging Using Mutual Information (RUMI), a method for online generation of robot action sequences to gather information about the pose of a known movable object in visually-occluded environments. Focusing on contact-rich rummaging, our approach leverages mutual information between the object pose distribution and robot trajectory for action planning. From an observed partial point cloud, RUMI deduces the compatible object pose distribution and approximates the mutual information of it with workspace occupancy in real time. Based on this, we develop an information gain cost function and a reachability cost function to keep the object within the robot's reach. These are integrated into a model predictive control (MPC) framework with a stochastic dynamics model, updating the pose distribution in a closed loop. Key contributions include a new belief framework for object pose estimation, an efficient information gain computation strategy, and a robust MPC-based control scheme. RUMI demonstrates superior performance in both simulated and real tasks compared to baseline methods.

Yubo Zhu, Dongrui Liu, Zecheng Lin et al.

Estimating the difficulty of input questions as perceived by large language models (LLMs) is essential for accurate performance evaluation and adaptive inference. Existing methods typically rely on repeated response sampling, auxiliary models, or fine-tuning the target model itself, which may incur substantial computational costs or compromise generality. In this paper, we propose a novel approach for difficulty estimation that leverages only the hidden representations produced by the target LLM. We model the token-level generation process as a Markov chain and define a value function to estimate the expected output quality given any hidden state. This allows for efficient and accurate difficulty estimation based solely on the initial hidden state, without generating any output tokens. Extensive experiments across both textual and multimodal tasks demonstrate that our method consistently outperforms existing baselines in difficulty estimation. Moreover, we apply our difficulty estimates to guide adaptive reasoning strategies, including Self-Consistency, Best-of-N, and Self-Refine, achieving higher inference efficiency with fewer generated tokens.

Sheng Zhong, Zhenyuan Zhang, Nima Fazeli et al.

We propose an approach to online model adaptation and control in the challenging case of hybrid and discontinuous dynamics where actions may lead to difficult-to-escape "trap" states, under a given controller. We first learn dynamics for a system without traps from a randomly collected training set (since we do not know what traps will be encountered online). These "nominal" dynamics allow us to perform tasks in scenarios where the dynamics matches the training data, but when unexpected traps arise in execution, we must find a way to adapt our dynamics and control strategy and continue attempting the task. Our approach, Trap-Aware Model Predictive Control (TAMPC), is a two-level hierarchical control algorithm that reasons about traps and non-nominal dynamics to decide between goal-seeking and recovery policies. An important requirement of our method is the ability to recognize nominal dynamics even when we encounter data that is out-of-distribution w.r.t the training data. We achieve this by learning a representation for dynamics that exploits invariance in the nominal environment, thus allowing better generalization. We evaluate our method on simulated planar pushing and peg-in-hole as well as real robot peg-in-hole problems against adaptive control, reinforcement learning, trap-handling baselines, where traps arise due to unexpected obstacles that we only observe through contact. Our results show that our method outperforms the baselines on difficult tasks, and is comparable to prior trap-handling methods on easier tasks.

Junjie Shi, Yuan Zhang, Sheng Zhong

Since the advent of software defined networks ({SDN}), there have been many attempts to outsource the complex and costly local network functionality, i.e. the middlebox, to the cloud in the same way as outsourcing computation and storage. The privacy issues, however, may thwart the enterprises' willingness to adopt this innovation since the underlying configurations of these middleboxes may leak crucial and confidential information which can be utilized by attackers. To address this new problem, we use firewall as an sample functionality and propose the first privacy preserving outsourcing framework and schemes in SDN. The basic technique that we exploit is a ground-breaking tool in cryptography, the \textit{cryptographic multilinear map}. In contrast to the infeasibility in efficiency if a naive approach is adopted, we devise practical schemes that can outsource the middlebox as a blackbox after \textit{obfuscating} it such that the cloud provider can efficiently perform the same functionality without knowing its underlying private configurations. Both theoretical analysis and experiments on real-world firewall rules demonstrate that our schemes are secure, accurate, and practical.