相关论文: Test-time Verification via Optimal Transport: Cove…

Connected vehicles (CVs) can provide numerous new data via vehicle-to-vehicle or vehicle-to-infrastructure communication. These data can in turn be used to facilitate real-time traffic state estimation. In this paper, we focus on ramp queue…

Uncertainty estimation is critical for deploying reasoning language models, yet remains poorly understood under extended chain-of-thought reasoning. We study parallel sampling as a fully black-box approach using verbalized confidence and…

Recent large language models (LLMs) achieve strong performance in generating promising reasoning paths for complex tasks. However, despite powerful generation ability, LLMs remain weak at verifying their own answers, revealing a persistent…

Self-supervised research improved greatly over the past half decade, with much of the growth being driven by objectives that are hard to quantitatively compare. These techniques include colorization, cyclical consistency, and…

Language models typically tokenize raw text into sequences of subword identifiers from a predefined vocabulary, a process inherently sensitive to typographical errors, length variations, and largely oblivious to the internal structure of…

As the complexity of control systems increases, the need for systematic methods to guarantee their efficacy grows as well. However, direct testing of these systems is oftentimes costly, difficult, or impractical. As a result, the test and…

Software testing and verification are critical for ensuring the reliability and security of modern software systems. Traditionally, formal verification techniques, such as model checking and theorem proving, have provided rigorous…

The availability of a wide range of large language models (LLMs) embedded in various agentic systems has significantly increased the potential of model selection strategies to improve the cost-performance tradeoff. Existing strategies…

We study the problem of entropy calibration, which asks whether a language model's entropy over generations matches its log loss on human text. Past work found that models are miscalibrated, with entropy per step increasing as generations…

As LLMs are deployed in high-stakes settings, users must judge the correctness of individual responses, often relying on model-generated justifications such as reasoning chains or explanations. Yet, no standard measure exists for whether…

We study the verification problem of stochastic systems under signal temporal logic (STL) specifications. We propose a novel approach that enables the verification of the probabilistic satisfaction of STL specifications for nonlinear…

This paper deals with a common verification methodology and environment for SystemC BCA and RTL models. The aim is to save effort by avoiding the same work done twice by different people and to reuse the same environment for the two design…

Most current results on coverage control using mobile sensors require that one partitioned cell is associated with precisely one sensor. In this paper, we consider a class of coverage control problems involving higher order Voronoi…

We examine the problem of learning sequential tasks from a single visual demonstration. A key challenge arises when demonstrations are temporally misaligned due to variations in timing, differences in embodiment, or inconsistencies in…

Test-time adaptation paradigm provides flexibility towards domain shifts by performing immediate adaptation on unlabeled target data from the source model. Vision-Language Models (VLMs) leverage their generalization capabilities for diverse…

Learning, prediction, and compression are intimately connected: a model that accurately predicts the next symbol in a sequence can be coupled with a source coder to compress that sequence near its information-theoretic limit. When tokenized…

Quantum optimal control experiments and simulations have successfully manipulated the dynamics of systems ranging from atoms to biomolecules. Surprisingly, these collective works indicate that the effort (i.e., the number of algorithmic…

Offline Reinforcement Learning (RL) presents a promising paradigm for training autonomous vehicle (AV) planning policies from large-scale, real-world driving logs. However, the extreme data imbalance in these logs, where mundane scenarios…

Verifiers have been demonstrated to enhance LLM reasoning via test-time scaling (TTS). Yet, they face significant challenges in complex domains. Error propagation from incorrect intermediate reasoning can lead to false positives for…

Large language models (LLMs) have shown promising performance across various tasks. However, their autoregressive decoding process poses significant challenges for efficient deployment on existing AI hardware. Quantization alleviates memory…