LLMs Know When They Know, but Do Not Act on It: A Metacognitive Harness for Test-time Scaling

Large language models (LLMs) often expose useful signals of self-monitoring: before solving a problem, they can estimate whether they are likely to succeed, and after solving it, they can judge whether their answer is likely to be correct. However, these signals are typically measured or elicited in isolation, rather than used to control inference. In this work, we ask whether LLMs possess latent metacognitive ability that can be turned into effective test-time control. Inspired by the Nelson--Narens theory from cognitive psychology, we propose a metacognitive harness that separates monitoring from reasoning. For each problem, the model first reports a pre-solve feeling-of-knowing (FOK) signal; after each solve attempt, it reports a post-solve judgment-of-learning (JOL) signal. Rather than treating these signals as passive confidence estimates, the harness turns them into an explicit control interface for reasoning: it decides when to trust the current solution, when to retry with compact metacognitive feedback, and when to pass multiple attempts to a final aggregator. Across text, code, and multimodal reasoning benchmarks, our harness substantially improves a fixed Claude Sonnet-4.6 base model without parameter updates or benchmark-specific fine-tuning. On the evaluated public benchmark snapshots, it raises pooled accuracy from 48.3 to 56.9 and exceeds the strongest listed leaderboard entries on the three primary evaluation settings: HLE-Verified, LiveCodeBench v6, and R-Bench-V. These results suggest that strong LLMs may already possess useful metacognitive ability, but require an explicit control harness to act on it during reasoning.

A Benchmark Grocery Dataset of Realworld Point Clouds From Single View

Fine-grained grocery object recognition is an important computer vision problem with broad applications in automatic checkout, in-store robotic navigation, and assistive technologies for the visually impaired. Existing datasets on groceries are mainly 2D images. Models trained on these datasets are limited to learning features from the regular 2D grids. While portable 3D sensors such as Kinect were commonly available for mobile phones, sensors such as LiDAR and TrueDepth, have recently been integrated into mobile phones. Despite the availability of mobile 3D sensors, there are currently no dedicated real-world large-scale benchmark 3D datasets for grocery. In addition, existing 3D datasets lack fine-grained grocery categories and have limited training samples. Furthermore, collecting data by going around the object versus the traditional photo capture makes data collection cumbersome. Thus, we introduce a large-scale grocery dataset called 3DGrocery100. It constitutes 100 classes, with a total of 87,898 3D point clouds created from 10,755 RGB-D single-view images. We benchmark our dataset on six recent state-of-the-art 3D point cloud classification models. Additionally, we also benchmark the dataset on few-shot and continual learning point cloud classification tasks. Project Page: https://bigdatavision.org/3DGrocery100/.

Unsupervised Out-of-Distribution Detection with Diffusion Inpainting

Unsupervised out-of-distribution detection (OOD) seeks to identify out-of-domain data by learning only from unlabeled in-domain data. We present a novel approach for this task - Lift, Map, Detect (LMD) - that leverages recent advancement in diffusion models. Diffusion models are one type of generative models. At their core, they learn an iterative denoising process that gradually maps a noisy image closer to their training manifolds. LMD leverages this intuition for OOD detection. Specifically, LMD lifts an image off its original manifold by corrupting it, and maps it towards the in-domain manifold with a diffusion model. For an out-of-domain image, the mapped image would have a large distance away from its original manifold, and LMD would identify it as OOD accordingly. We show through extensive experiments that LMD achieves competitive performance across a broad variety of datasets.

A principled distributional approach to trajectory similarity measurement

Existing measures and representations for trajectories have two longstanding fundamental shortcomings, i.e., they are computationally expensive and they can not guarantee the `uniqueness' property of a distance function: dist(X,Y) = 0 if and only if X=Y, where $X$ and $Y$ are two trajectories. This paper proposes a simple yet powerful way to represent trajectories and measure the similarity between two trajectories using a distributional kernel to address these shortcomings. It is a principled approach based on kernel mean embedding which has a strong theoretical underpinning. It has three distinctive features in comparison with existing approaches. (1) A distributional kernel is used for the very first time for trajectory representation and similarity measurement. (2) It does not rely on point-to-point distances which are used in most existing distances for trajectories. (3) It requires no learning, unlike existing learning and deep learning approaches. We show the generality of this new approach in three applications: (a) trajectory anomaly detection, (b) anomalous sub-trajectory detection, and (c) trajectory pattern mining. We identify that the distributional kernel has (i) a unique data-dependent property and the above uniqueness property which are the key factors that lead to its superior task-specific performance; and (ii) runtime orders of magnitude faster than existing distance measures.