While Diffusion Large Language Models (dLLMs) offer structural advantages for global planning, efficiently verifying that they arrive at correct answers via valid reasoning traces remains a critical challenge. In this work, we propose a geometric perspective: Reasoning on the Manifold. We hypothesize that valid generat...
Jiaoyang Ruan, Xin Gao, Yinda Chen et al.· 0 citations
Looped transformers promise test-time compute scaling by spending more iterations on harder problems, but it remains unclear which architectural choices let them extrapolate to harder problems at test time rather than memorize training-specific solutions. We introduce a fixed-point based framework for analyzing looped...
The Forward-Forward (FF) algorithm trains networks layer-by-layer using a local "goodness function," yet the choice of activity statistic remains an important design question. We systematically study how selectivity and activity shape affect FF learning, introducing selective goodness functions based on top-$k$ and ent...
Talha Ruzgar Akkus, Suayp Talha Kocabay, Kamer Ali Yuksel et al.· 0 citations
We show that robustness to post-training quantization (PTQ) is a transferable direction in weight space. We call this direction the quantization vector: extracted from a donor task by simple weight-space arithmetic, it can be used to patch a receiver model and improve PTQ Top-1 accuracy by up to 60% in a 3-bit setting,...
Daniele Solombrino, Antonio Andrea Gargiulo, Alessandro Zirilli et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Explainable AI (XAI) methods are commonly evaluated using functional correctness metrics, also called faithfulness or fidelity, which estimate how closely an explanation reflects the model's reasoning. Higher correctness is assumed to produce better human understanding, but this link has not been tested with controlled...
Gregor Baer, Chao Zhang, Isel Grau et al.· 0 citations
Tactile vision-language-action models (VLAs) and tactile world-action models (WAMs) are a promising new paradigm for contact-rich manipulation, but existing models either underutilize touch, rely on large-scale tactile pretraining, or incur high inference latency due to iterative video generation. We present VTAM, an e...
Haoran Yuan, Weigang Yi, Zhenyu Zhang et al.· 0 citations
As one of the most important underwater sensing technologies, forward-looking sonar exhibits unique imaging characteristics. Sonar images are often affected by severe speckle noise, low texture contrast, acoustic shadows, and geometric distortions. These factors make it difficult for traditional teacher-student framewo...
Ping Guo, Chengzhou Li, Guanchen Meng et al.· 0 citations
Low-light image enhancement (LLIE) supports optical sensing and scene observation when limited photon capture and sensor noise degrade image quality. However, existing methods can introduce colour deviations or blur local details in the enhanced images. To address this challenge, we propose CueDR-Net, a dual-path netwo...
Yicui Shi, Yuhan Chen, Xiangfei Huang et al.· 0 citations
Deploying autonomous driving systems requires robustness against long-tail scenarios that are rare but safety-critical. While adversarial training offers a promising solution, existing methods typically decouple scenario generation from policy optimization and rely on heuristic surrogates. This leads to objective misal...
Tong Nie, Yihong Tang, Junlin He et al.· 0 citations
The Segment Anything Model (SAM) has revolutionized interactive segmentation through spatial prompting. While existing work primarily focuses on automating prompts in various settings, real-world annotation workflows involve iterative refinement where annotators observe model outputs and strategically place prompts to...
Animal re-identification (ReID) faces critical challenges due to viewpoint variations, particularly in Aerial-Ground (AG-ReID) settings where models must match individuals across drastic elevation changes. However, existing datasets lack the precise angular annotations required to systematically analyze these geometric...
William Grolleau, Achraf Chaouch, Astrid Sabourin et al.· 0 citations
Trainable-frequency quantum models train the prefactors of their encoding gates so that the circuit's frequencies can move onto the target's. From the unit start of prior work, training rarely reaches a target more than one resolution step away. We show that the target-driven part of the prefactor gradient is a weighte...
Michael Poppel, Markus Baumann, Sebastian W\"olckert et al.· 0 citations
