Large language models (LLMs) have shown increasing capability in problemsolving and decision-making, largely based on the step-by-step chain-of-thought reasoning processes. However, evaluating these reasoning abilities has become increasingly challenging. Existing outcome-based benchmarks are beginning to saturate, becoming less effective in tracking meaningful progress. To address this, we present a process-based benchmark MR-Ben that demands a meta-reasoning skill, where LMs are asked to locate and analyse potential errors in automatically generated reasoning steps. Our meta-reasoning paradigm is especially suited for system-2 slow thinking, mirroring the human cognitive process of carefully examining assumptions, conditions, calculations, and logic to identify mistakes. MR-Ben comprises 5,975 questions curated by human experts across a wide range of subjects, including physics, chemistry, logic, coding, and more. Through our designed metrics for assessing meta-reasoning on this benchmark, we identify interesting limitations and weaknesses of current LLMs (open-source and closed-source models).

In this technical supplement, we provide detailed insights and additional results to support our main paper. Section A.1 outlines the generation process of the SynRS3D dataset, including the tools and plugins used. It also covers the licenses for these plugins. Section A.3 elaborates on the evaluation metrics for different tasks, including the proposed F Section A.4 describes the experimental setup and the selection of hyperparameters for the RS3DAda method. Section A.5 presents the ablation study results and analysis for the RS3DAda method. Section A.6 provides supplementary experimental results combining SynRS3D and real data scenarios, complementing Section 5.2 of the main paper. Section A.9 highlights the performance of models trained on the SynRS3D dataset using RS3DAda in the critical application of disaster mapping in remote sensing. A.1 Detailed Generation Workflow of SynRS3D The generation workflow of SynRS3D involves several key steps, from initializing sensor and sunlight parameters to generating the layout, geometry, and textures of the scene. This comprehensive process ensures that the generated SynRS3D mimics real-world remote sensing scenarios with high fidelity. The main steps of the workflow are as follows: Initialization: Set up the sensor and sunlight parameters using uniform and normal distributions to simulate various conditions. Layout Generation: Define the grid and terrain parameters to create diverse urban and natural environments. Texture Generation: Use advanced models like GPT-4 [1] and Stable Diffusion [18] to generate realistic textures for different categories of land cover.

Existing ML benchmarks lack the depth and diversity of annotations needed for evaluating models on business process management (BPM) tasks. BPM is the practice of documenting, measuring, improving, and automating enterprise workflows. However, research has focused almost exclusively on one task - full end-to-end automation using agents based on multimodal foundation models (FMs) like GPT-4. This focus on automation ignores the reality of how most BPM tools are applied today - simply documenting the relevant workflow takes 60% of the time of the typical process optimization project.

Text-guided diffusion models have significantly advanced image editing, enabling high-quality and diverse modifications driven by text prompts. However, effective editing requires inverting the source image into a latent space, a process often hindered by prediction errors inherent in DDIM inversion. These errors accumulate during the diffusion process, resulting in inferior content preservation and edit fidelity, especially with conditional inputs. We address these challenges by investigating the primary contributors to error accumulation in DDIM inversion and identify the singularity problem in traditional noise schedules as a key issue. To resolve this, we introduce the Logistic Schedule, a novel noise schedule designed to eliminate singularities, improve inversion stability, and provide a better noise space for image editing. This schedule reduces noise prediction errors, enabling more faithful editing that preserves the original content of the source image. Our approach requires no additional retraining and is compatible with various existing editing methods. Experiments across eight editing tasks demonstrate the Logistic Schedule's superior performance in content preservation and edit fidelity compared to traditional noise schedules, highlighting its adaptability and effectiveness.

We consider sequential treatment regimes where each unit is exposed to combinations of interventions over time. When interventions are described by qualitative labels, such as "close schools for a month due to a pandemic" or "promote this podcast to this user during this week", it is unclear which appropriate structural assumptions allow us to generalize behavioral predictions to previously unseen combinations of interventions. Standard black-box approaches mapping sequences of categorical variables to outputs are applicable, but they rely on poorly understood assumptions on how reliable generalization can be obtained, and may underperform under sparse sequences, temporal variability, and large action spaces. To approach that, we pose an explicit model for composition, that is, how the effect of sequential interventions can be isolated into modules, clarifying which data conditions allow for the identification of their combined effect at different units and time steps. We show the identification properties of our compositional model, inspired by advances in causal matrix factorization methods. Our focus is on predictive models for novel compositions of interventions instead of matrix completion tasks and causal effect estimation. We compare our approach to flexible but generic black-box models to illustrate how structure aids prediction in sparse data conditions.

Breaking down a problem into intermediate steps has demonstrated impressive performance in Large Language Model (LLM) reasoning. However, the growth of the reasoning chain introduces uncertainty and error accumulation, making it challenging to elicit accurate final results. To tackle this challenge of uncertainty in multi-step reasoning, we introduce a stepwise self-evaluation mechanism to guide and calibrate the reasoning process of LLMs. We propose a decoding algorithm integrating the self-evaluation guidance via stochastic beam search. The selfevaluation guidance serves as a better-calibrated automatic criterion, facilitating an efficient search in the reasoning space and resulting in superior prediction quality.

This paper presents the official release of the Digital Typhoon dataset, the longest typhoon satellite image dataset for 40+ years aimed at benchmarking machine learning models for long-term spatio-temporal data. To build the dataset, we developed a workflow to create an infrared typhoon-centered image for cropping using Lambert azimuthal equal-area projection referring to the best track data. We also address data quality issues such as inter-satellite calibration to create a homogeneous dataset. To take advantage of the dataset, we organized machine learning tasks by the types and targets of inference, with other tasks for meteorological analysis, societal impact, and climate change. The benchmarking results on the analysis, forecasting, and reanalysis for the intensity suggest that the dataset is challenging for recent deep learning models, due to many choices that affect the performance of various models. This dataset reduces the barrier for machine learning researchers to meet large-scale real-world events called tropical cyclones and develop machine learning models that may contribute to advancing scientific knowledge on tropical cyclones as well as solving societal and sustainability issues such as disaster reduction and climate change.

Practical online learning tasks are often naturally defined on unconstrained domains, where optimal algorithms for general convex losses are characterized by the notion of comparator adaptivity. In this paper, we design such algorithms in the presence of switching cost - the latter penalizes the typical optimism in adaptive algorithms, leading to a delicate design trade-off. Based on a novel dual space scaling strategy discovered by a continuous-time analysis, we propose a simple algorithm that improves the existing comparator adaptive regret bound [ZCP22a] to the optimal rate. The obtained benefits are further extended to the expert setting, and the practicality of the proposed algorithm is demonstrated through a sequential investment task.