Large Reasoning Models (LRMs) have recently achieved remarkable success in complex reasoning tasks.However, closer scrutiny reveals persistent failure modes compromising performance and cost: I) Intra-step level, marked by calculation or derivation errors; II) Inter-step level, involving oscillation and stagnation; and III) Instance level, causing maladaptive over-thinking.Existing endeavors target isolated levels without unification, while their black-box nature and reliance on RL hinder explainability and controllability.To bridge these gaps, we conduct an in-depth white-box analysis, identifying key neurons (Mixture of Neurons, MoN) and their fluctuation patterns associated with distinct failures.Building upon these insights, we propose NeuReasoner, an explainable, controllable, and unified reasoning framework driven by MoN.Technically, NeuReasoner integrates lightweight MLPs for failure detection with a special token-triggered self-correction mechanism learned via SFT.During inference, special tokens are inserted upon failure detection to actuate controllable remedial behaviors.Extensive evaluations across six benchmarks, six backbone models (8B∼70B) against nine competitive baselines, demonstrate that NeuReasoner achieves performance gains of up to 27.0% while reducing token consumption by 19.6% ∼ 63.3%.... Correct Answer...

Recent Large Reasoning Models (LRMs) likeDeepSeek-R1 have demonstrated remarkable success in complex reasoning tasks, exhibiting human-like patterns in exploring multiple alternative solutions.Upon closer inspection, however, we uncover a surprising phenomenon: The First is The Best, where alternative solutions are not merely suboptimal but potentially detrimental.This observation challenges widely accepted test-time scaling laws, leading us to hypothesize that errors within the reasoning path scale concurrently with test time.Through comprehensive empirical analysis, we characterize errors as a forest-structured Forest of Errors (FoE) and conclude that FoE makes the First the Best, which is underpinned by rigorous theoretical analysis.Leveraging these insights, we propose RED, a self-guided efficient reasoning framework comprising two components: I) Refining First, which suppresses FoE growth in the first solution; and II) Discarding Subs, which prunes subsequent FoE via dualconsistency.Extensive experiments across five benchmarks and six backbone models demonstrate that RED outperforms eight competitive baselines, achieving performance gains of up to 19.0% while reducing token consumption by 37.7% ∼ 70.4%.Moreover, comparative experiments on FoE metrics shed light on how RED achieves effectiveness.... ... ...

Graph Neural Networks (GNNs) have shown remarkable capability in modeling complex relational data, yet their effectiveness is constrained by subtle but critical limitations in the message-passing paradigm. We identify a fundamental issue -- termed aggregation collapse -- arising from the aggregation step that combines representations of central and neighboring nodes with varying sizes. This bottleneck manifests in two forms of inherent information loss: (i) feature information loss, occurring when value distributions in learned representations become concentrated and indistinguishable after aggregation, and (ii) structure information loss, where variable node degrees are obscured by aggregation functions. To address these challenges, we propose complementary solutions: (i) nonlinear feature mapping and distribution re-scaler, which diversify input feature distributions prior to aggregation to preserve valuable feature-level details, and (ii) post-aggregation structural encoding, which retains essential structure information. Our methods are theoretically grounded, easy to implement, and broadly applicable to existing GNN models. Empirical evaluations demonstrate that our techniques successfully alleviate aggregation collapse, leading to improved performance and scalability in GNNs.

Low-Rank Adaptation (LoRA) is a widely adopted parameter-efficient fine-tuning (PEFT) method validated across NLP and CV domains. However, LoRA faces an inherent low-rank bottleneck: narrowing its performance gap with full fine-tuning requires increasing the rank of its parameter matrix, resulting in significant parameter overhead. Recent linear LoRA variants have attempted to enhance expressiveness by introducing additional linear mappings; however, their composition remains inherently linear and fails to fundamentally improve LoRA’s representational capacity. To address this limitation, we propose \ourmethod, which incorporates an Adaptive Nonlinear Layer (ANL) between two linear projectors to capture \emph{fixed} and \emph{learnable} nonlinearities. This combination forms an {\fontfamily{lmtt}\selectfont \textbf{MLP-like structure}} with a compressed rank, enabling flexible and precise approximation of diverse target functions while theoretically guaranteeing lower approximation errors and bounded gradients. Extensive experiments on 22 datasets and 6 pretrained models demonstrate that \ourmethod: (\textbf{I}) not only matches or surpasses full fine-tuning performance with only $6.18\%\sim25\%$ of LoRA’s parameters but also (\textbf{II}) outperforms state-of-the-art PEFT methods by up to $10.88\%$ in both NLP and CV tasks, and \textbf{(III)} exhibits robust performance across various rank configurations.

Training high-quality deep models necessitates vast amounts of data, resulting in overwhelming computational and memory demands. Recently, data pruning, distillation, and coreset selection have been developed to streamline data volume by \textit{retaining}, \textit{synthesizing}, or \textit{selecting} a small yet informative subset from the full set. Among these methods, data pruning incurs the least additional training cost and offers the most practical acceleration benefits. However, it is the most vulnerable, often suffering significant performance degradation with imbalanced or biased data schema, thus raising concerns about its accuracy and reliability in on-device deployment. Therefore, there is a looming need for a new data pruning paradigm that maintains the efficiency of previous practices while ensuring balance and robustness. Unlike the fields of computer vision and natural language processing, where mature solutions have been developed to address these issues, graph neural networks (GNNs) continue to struggle with increasingly large-scale, imbalanced, and noisy datasets, lacking a unified dataset pruning solution. To achieve this, we introduce a novel dynamic soft-pruning method, \ourmethod, designed to update the training ``basket'' during the process using trainable prototypes. \ourmethod first constructs a well-modeled graph embedding hypersphere and then samples \textit{representative, balanced, and unbiased subsets} from this embedding space, which achieves the goal we called {\fontfamily{lmtt}\selectfont \textbf{Graph Training Debugging}}. Extensive experiments on four datasets across three GNN backbones, demonstrate that \ourmethod (I) achieves or surpasses the performance of the full dataset with $30\%\sim50\%$ fewer training samples, (II) attains up to a $2.81\times$ lossless training speedup, and (III) outperforms state-of-the-art pruning methods in imbalanced training and noisy training scenarios by $0.3\%\sim4.3\%$ and $3.6\%\sim7.8\%$, respectively.