Search

Divergent thinking, the cognitive process of generating diverse solutions, is a hallmark of human creativity and problem-solving. For machines, sampling diverse solution trajectories in complex reasoning problems is crucial for robust outcomes, data augmentation, and enhanced model generalization. Large language models (LLMs) often struggle with generating high-quality, diverse reasoning. While supervised fine-tuning helps with quality, it requires extensive supervision data to capture the full diversity of solutions. Alternatively, reinforcement learning methods like PPO aim to find limited highest-reward solutions while neglecting the solution diversity, akin to convergent thinking. To address these limitations, we propose Flow of Reasoning (FoR) -- an efficient LLM training approach enabling diverse reasoning with minimal data. FoR formulates multi-step LLM reasoning as a Markovian flow from an initial state to terminal states. The formulation allows to adapt principled GFlowNet approaches to train the LLM as a policy, which is able to sample multiple reasoning paths with probabilities proportional to the unnormalized reward. Empirical results show that, with limited training data (e.g., 15 examples), FoR can discover diverse high-quality solutions that excel greatly beyond current state-of-the-art methods across three tasks, including embodied reasoning (BlocksWorld), math puzzle solving (Game24), and logical reasoning (PrOntoQA). Code is available at https://github.com/Yu-Fangxu/FoR.

The abdominal electrocardiogram (AECG) gives a safe and non-invasive way to monitor fetal well-being during pregnancy. However, due to the overlap with maternal ECG (MECG) as well as significant external noise, it is challenging to extract weak fetal ECG (FECG) using surface electrodes. In this study, we introduce a novel periodic progressive FastICA peel-off (PPFP) method for noninvasive extraction of weak surface FECG signals, leveraging the two-step FastICA method and a peel-off strategy from the progressive FastICA peel-off (PFP) approach. Specifically, for ECG signals, the periodic constrained FastICA that integrates ECG signal characteristics enables precise extraction of MECG and FECG spike trains. Additionally, a peel-off strategy incorporating SVD waveform reconstruction ensures comprehensive identification of subtle source signals. The performance of the proposed method was examined on public datasets with reference, synthetic data and clinical data, with an F1-scores for FECG extraction on public dataset of 99.59%, on synthetic data with the highest noise level of 99.50%, which are all superior to other comparative methods. Furthermore, clearly periodic and physiologically consistent FECG signals were extracted from clinically collected data. The results indicates that our proposed method has potential and effectiveness to separate MECG and weak FECG from multichannel AECG with high precision in high noise condition, which is of vital importance for ensuring the safety of both the fetus and the mother, as well as the advancement of artificial intelligent clinical monitoring.

Drama is a form of storytelling inspired by human creativity, proceeding with a predefined storyline, carrying emotions and thoughts. This paper introduces \emph{LLM-based interactive drama}, which endows traditional drama with an unprecedented immersion, where a person is allowed to walk into it and interact with the characters and scenes. We define this new artistic genre by 6 essential elements-plot, character, thought, diction, spectacle and interaction-and study the entire pipeline to forge a backbone \emph{drama LLM} to drive the playing process, which is challenged by limited drama resources, uncontrollable narrative development, and complicated instruction following. We propose \emph{Narrative Chain} to offer finer control over the narrative progression during interaction with players; \emph{Auto-Drama} to synthesize drama scripts given arbitrary stories; \emph{Sparse Instruction Tuning} to allow the model to follow sophisticated instructions. We manually craft 3 scripts, \emph{Detective Conan}, \emph{Harry Potter}, \emph{Romeo and Juliet}, and design a 5-dimension principle to evaluate the drama LLM comprehensively., Comment: Accepted by ACL 2024 Findings

In this paper, we first present a simple lemma which allows us to estimate the box dimension of graphs of given functions by the associated oscillation sums and oscillation vectors. Then we define vertical scaling matrices of generalized affine fractal interpolation surfaces (FISs). By using these matrices, we establish relationships between oscillation vectors of different levels, which enables us to obtain the box dimension of generalized affine FISs under certain constraints.

Let $f$ be a generalized affine fractal interpolation function with vertical scaling functions. In this paper, we prove the monotonicity of spectral radii of vertical scaling matrices without additional assumptions. We also obtain the irreducibility of these matrices under certain conditions. By these results, we estimate $\mathrm{dim}_B \Gamma f$, the box dimension of the graph of $f$, by the limits of spectral radii of vertical scaling matrices. We also estimate $\mathrm{dim}_B \Gamma f$ directly by the sum function of vertical scaling functions. As an application, we study the box dimension of the graph of a generalized Weierstrass-type function., Comment: 20 pages, 1 figure. arXiv admin note: text overlap with arXiv:2208.04126

Correlation based stereo matching has achieved outstanding performance, which pursues cost volume between two feature maps. Unfortunately, current methods with a fixed model do not work uniformly well across various datasets, greatly limiting their real-world applicability. To tackle this issue, this paper proposes a new perspective to dynamically calculate correlation for robust stereo matching. A novel Uncertainty Guided Adaptive Correlation (UGAC) module is introduced to robustly adapt the same model for different scenarios. Specifically, a variance-based uncertainty estimation is employed to adaptively adjust the sampling area during warping operation. Additionally, we improve the traditional non-parametric warping with learnable parameters, such that the position-specific weights can be learned. We show that by empowering the recurrent network with the UGAC module, stereo matching can be exploited more robustly and effectively. Extensive experiments demonstrate that our method achieves state-of-the-art performance over the ETH3D, KITTI, and Middlebury datasets when employing the same fixed model over these datasets without any retraining procedure. To target real-time applications, we further design a lightweight model based on UGAC, which also outperforms other methods over KITTI benchmarks with only 0.6 M parameters., Comment: Accepted by ICCV2023

In recent years, deep learning has made significant progress in wood panel defect detection. However, there are still challenges such as low detection , slow detection speed, and difficulties in deploying embedded devices on wood panel surfaces. To overcome these issues, we propose a lightweight wood panel defect detection method called YOLOv5-LW, which incorporates attention mechanisms and a feature fusion network.Firstly, to enhance the detection capability of acceptable defects, we introduce the Multi-scale Bi-directional Feature Pyramid Network (MBiFPN) as a feature fusion network. The MBiFPN reduces feature loss, enriches local and detailed features, and improves the model's detection capability for acceptable defects.Secondly, to achieve a lightweight design, we reconstruct the ShuffleNetv2 network model as the backbone network. This reconstruction reduces the number of parameters and computational requirements while maintaining performance. We also introduce the Stem Block and Spatial Pyramid Pooling Fast (SPPF) models to compensate for any accuracy loss resulting from the lightweight design, ensuring the model's detection capabilities remain intact while being computationally efficient.Thirdly, we enhance the backbone network by incorporating Efficient Channel Attention (ECA), which improves the network's focus on key information relevant to defect detection. By attending to essential features, the model becomes more proficient in accurately identifying and localizing defects.We validate the proposed method using a self-developed wood panel defect dataset.The experimental results demonstrate the effectiveness of the improved YOLOv5-LW method. Compared to the original model, our approach achieves a 92.8\% accuracy rate, reduces the number of parameters by 27.78\%, compresses computational volume by 41.25\%, improves detection inference speed by 10.16\%

Let $f$ be a generalized affine fractal interpolation function with vertical scaling function $S$. In this paper, we study $\dim_B \Gamma f$, the box dimension of the graph of $f$, under the assumption that $S$ is a Lipschtz function. By introducing vertical scaling matrices, we estimate the upper bound and the lower bound of oscillations of $f$. As a result, we obtain explicit formula of $\dim_B \Gamma f$ under certain constraint conditions., Comment: 18 pages, 1 figure

As a widely studied task, video restoration aims to enhance the quality of the videos with multiple potential degradations, such as noises, blurs and compression artifacts. Among video restorations, compressed video quality enhancement and video super-resolution are two of the main tacks with significant values in practical scenarios. Recently, recurrent neural networks and transformers attract increasing research interests in this field, due to their impressive capability in sequence-to-sequence modeling. However, the training of these models is not only costly but also relatively hard to converge, with gradient exploding and vanishing problems. To cope with these problems, we proposed a two-stage framework including a multi-frame recurrent network and a single-frame transformer. Besides, multiple training strategies, such as transfer learning and progressive training, are developed to shorten the training time and improve the model performance. Benefiting from the above technical contributions, our solution wins two champions and a runner-up in the NTIRE 2022 super-resolution and quality enhancement of compressed video challenges. Code is available at https://github.com/ryanxingql/winner-ntire22-vqe., Comment: Winning two championships and one runner-up in the NTIRE 2022 challenge on super-resolution and quality enhancement of compressed video; Accepted to CVPRW 2022

This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video. In this challenge, we proposed the LDV 2.0 dataset, which includes the LDV dataset (240 videos) and 95 additional videos. This challenge includes three tracks. Track 1 aims at enhancing the videos compressed by HEVC at a fixed QP. Track 2 and Track 3 target both the super-resolution and quality enhancement of HEVC compressed video. They require x2 and x4 super-resolution, respectively. The three tracks totally attract more than 600 registrations. In the test phase, 8 teams, 8 teams and 12 teams submitted the final results to Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of super-resolution and quality enhancement of compressed video. The proposed LDV 2.0 dataset is available at https://github.com/RenYang-home/LDV_dataset. The homepage of this challenge (including open-sourced codes) is at https://github.com/RenYang-home/NTIRE22_VEnh_SR.

We point out a new type of diurnal effect for the cosmic ray boosted dark matter (DM). The DM-nucleon interactions not only allow the direct detection of DM with nuclear recoils, but also allow cosmic rays to scatter with and boost the nonrelativistic DM to higher energies. If the DM-nuclei scattering cross sections are sufficiently large, the DM flux is attenuated as it propagates through the Earth, leading to a strong diurnal modulation. This diurnal modulation provides another prominent signature for the direct detection of boosted sub-GeV DM, in addition to signals with higher recoil energy., Comment: 10 pages, 8 figures [with Supplementary Material], published in PRL