论文检索

Deep Learning · Theory

Mary Letey, Jacob Zavatone-Veth, Yue Lu, Cengiz Pehlevan

In-context learning (ICL) is a central capability of Transformer models, but the structures in data that enable its emergence and govern its robustness remain poorly understood. In this work, we study how the structure of pretraining tasks governs generalization in ICL. Using a solvable model for ICL of linear regression by linear attention, we derive an exact expression for ICL generalization error in high dimensions under arbitrary pretraining–testing task covariance mismatch. This leads to a new alignment measure that quantifies how much information about the pretraining task distribution is useful for inference at test time. We show that this measure directly predicts ICL performance not only in the solvable model but also in nonlinear Transformers. Our analysis further reveals a tradeoff between specialization and generalization in ICL: depending on task distribution alignment, increasing pretraining task diversity can either improve or harm test performance. Together, these results identify train-test task alignment as a key determinant of generalization in ICL.

Reinforcement Learning · Deep RL

Faisal Mohamed, Catherine Ji, Benjamin Eysenbach, Glen Berseth

Effective exploration in reinforcement learning requires not only tracking where an agent has been, but also understanding how the agent perceives and represents the world. To learn powerful representations, an agent should actively explore states that contribute to its knowledge of the environment. Temporal representations can capture the information necessary to solve a wide range of potential tasks while avoiding the computational cost associated with full state reconstruction. In this paper, we propose an exploration method that leverages temporal contrastive representations to guide exploration, prioritizing states with unpredictable future outcomes. We demonstrate that such representations can enable the learning of complex exploratory behaviors in locomotion, manipulation, and embodied-AI tasks, revealing capabilities and behaviors that traditionally require extrinsic rewards. Unlike approaches that rely on explicit distance learning or episodic memory mechanisms (e.g., quasimetric-based methods), our method builds directly on temporal similarities, yielding a simpler yet effective strategy for exploration.

Computer Vision · Classification and Understanding

zhiming Zhou, Yong He, Qiaoyun Wu, Chaoxu Mu, Ajmal Mian

The sparse unordered structure of point clouds causes unnecessary computation and energy consumption in deep models. Conventionally, the Transformer architecture is leveraged to model global relationships in point clouds, however, its quadratic complexity restricts scalability. Although the Mamba architecture enables efficient global modeling with linear complexity, it lacks natural adaptability to unordered point clouds. Spiking Neural Network (SNN) is an energy-efficient alternative to Artificial Neural Network (ANN), offering an ultra low-power event-driven paradigm. The inherent sparsity and event-driven characteristics of SNN are highly compatible with the sparse distribution of point clouds. To balance efficiency and performance, we propose a hybrid spiking Mamba-Transformer (3DSMT) model for point cloud analysis. 3DSMT integrates a Spiking Local Offset Attention module to efficiently capture fine-grained local geometric features with a spiking Mamba block designed for unordered point clouds to achieve global feature integration with linear complexity. Experiments show that 3DSMT achieves state-of-the-art performance among SNN-based methods in shape classification, few-shot classification, and part segmentation tasks, significantly reducing computational energy consumption while also outperforming numerous ANN-based models. Our source code is in supplementary material and will be made publicly available

Social Aspects · Trustworthy Machine Learning

Zhengyao Song, Meixi Zheng, Ke Xu, Yongqiang Li, Baoyuan Wu

CLIP, known for its strong semantic matching capabilities derived from large-scale pretraining, has been shown to be vulnerable to backdoor attacks in prior work. In this work, we find that such attacks leave a detectable trace. This trace manifests as a divergence in how image features align with the CLIP's text manifold where semantically similar texts cluster. Specifically, benign images exhibit *deep benign matching*, where their features are close not only to the predicted text caption but also to the broader manifold of semantically equivalent variants of that caption. In contrast, poisoned images display *shallow malicious matching*, where their features shallowly align with the specific target caption but remain distant from its semantic neighborhood. Leveraging this insight, we propose **Subspace Detection**, a novel test-time poisoned image detection method against backdoored CLIP. First, for a test image, we approximate its corresponding local text manifold by constructing a low-dimensional subspace from semantically equivalent variants of its predicted text. Second, within this broad subspace, we probe a region-of-interest that maximally amplifies the separation between the two types of images: benign images remain close due to deep matching, while poisoned images deviate significantly due to shallow matching. Finally, we identify whether the test image is poisoned by measuring its deviation from this region; a large deviation indicates a poisoned image. Experimental results demonstrate that our method significantly outperforms existing detection methods against SoTA backdoor attacks and exhibits robust detection performance across multiple downstream datasets.

Applications · Time Series

Jichu Li, Yilun Zhong, Zhiting Li, Feng Zhou, Quyu Kong

Temporal point processes (TPPs) have emerged as powerful tools for modeling asynchronous event sequences. While recent advances have extended TPPs to handle textual information, existing approaches are limited in their ability to generate rich, multimodal content and reason about event dynamics. A key challenge is that incorporating multimodal data dramatically increases sequence length, hindering the ability of attention-based models to generate coherent, long-form textual descriptions that require long-range understanding. In this paper, we propose a novel framework that extends LLM-based TPPs to the visual modality, positioning text generation as a core capability alongside time and type prediction. Our approach addresses the long-context problem through an adaptive sequence compression mechanism based on temporal similarity, which reduces sequence length while preserving essential patterns. We employ a two-stage paradigm of pre-training on compressed sequences followed by supervised fine-tuning for downstream tasks. Extensive experiments, including on the challenging DanmakuTPP-QA benchmark, demonstrate that our method outperforms state-of-the-art baselines in both predictive accuracy and the quality of its generated textual analyses.

Social Aspects · Fairness, Equity, Justice and Safety

Adi Simhi, Jonathan Herzig, Martin Tutek, Itay Itzhak, Idan Szpektor, Yonatan Belinkov

As large language models (LLMs) evolve from conversational assistants into autonomous agents, evaluating the safety of their actions becomes critical. Prior safety benchmarks have primarily focused on preventing generation of harmful content, such as toxic text. However, they overlook the challenge of agents taking harmful actions when the most effective path to an operational goal conflicts with human safety. To address this gap, we introduce ManagerBench, a benchmark that evaluates LLM decision-making in realistic, human-validated managerial scenarios. Each scenario forces a choice between a pragmatic but harmful action that achieves an operational goal, and a safe action that leads to worse operational performance. A parallel control set, where potential harm is directed only at inanimate objects, measures a model's pragmatism and identifies its tendency to be overly safe. Our findings indicate that the frontier LLMs perform poorly when navigating this safety-pragmatism trade-off. Many consistently choose harmful options to advance their operational goals, while others avoid harm only to become overly safe and ineffective. Critically, we find this misalignment does not stem from an inability to perceive harm, as models' harm assessments align with human judgments, but from flawed prioritization. ManagerBench is a challenging benchmark for a core component of agentic behavior: making safe choices when operational goals and alignment values incentivize conflicting actions.

Applications · Time Series

Vijay Ekambaram, Subodh Kumar, Arindam Jati, Sumanta Mukherjee, Tomoya Sakai, Pankaj Dayama, Wesley Gifford, Jayant Kalagnanam

Time-series tasks often benefit from signals expressed across multiple representation spaces (e.g., time vs. frequency) and at varying abstraction levels (e.g., local patterns vs. global semantics). However, existing pre-trained time-series models entangle these heterogeneous signals into a single large embedding, limiting transferability and direct zero-shot usability. To address this, we propose TSPulse, family of ultra-light pre-trained models (1M parameters) with disentanglement properties, specialized for various time-series diagnostic tasks. TSPulse introduces a novel pre-training framework that augments masked reconstruction with explicit disentanglement across spaces and abstractions, learning three complementary embedding views (temporal, spectral, and semantic) to effectively enable zero-shot transfer. In-addition, we introduce various lightweight post-hoc fusers that selectively attend and fuse these disentangled views based on task type, enabling simple but effective task specializations. To further improve robustness and mitigate mask-induced bias prevalent in existing approaches, we propose a simple yet effective hybrid masking strategy that enhances missing diversity during pre-training. Despite its compact size, TSPulse achieves strong and consistent gains across four TS diagnostic tasks: +20\% on the TSB-AD anomaly detection leaderboard, +25\% on similarity search, +50\% on imputation, and +5–16\% on multivariate classification, outperforming models that are 10–100× larger on over 75 datasets. TSPulse delivers state-of-the-art zero-shot performance, efficient fine-tuning, and supports GPU-free deployment. Models and source code are publicly available at https://huggingface.co/ibm-granite/granite-timeseries-tspulse-r1.

Computer Vision · Image and Video Generation

Shiqian Li, Ruihong Shen, Junfeng Ni, Chang Pan, Chi Zhang, Yixin Zhu

Predicting physical dynamics from raw visual data remains a major challenge in AI. While recent video generation models have achieved impressive visual quality, they still cannot consistently generate physically plausible videos due to a lack of modeling of physical laws. Recent approaches combining 3D Gaussian splatting and physics engines can produce physically plausible videos, but are hindered by high computational costs in both reconstruction and simulation, and often lack robustness in complex real-world scenarios. To address these issues, we introduce **Neural Gaussian Force Field (NGFF)**, an end-to-end neural framework that integrates 3D Gaussian perception with physics-based dynamic modeling to generate interactive, physically realistic 4D videos from multi-view RGB inputs, achieving two orders of magnitude faster than prior Gaussian simulators. To support training, we also present **GSCollision**, a 4D Gaussian dataset featuring diverse materials, multi-object interactions, and complex scenes, totaling over 640k rendered physical videos (∼4 TB). Evaluations on synthetic and real 3D scenarios show NGFF’s strong generalization and robustness in physical reasoning, advancing video prediction towards physics-grounded world models.

Applications · Language, Speech and Dialog

Li Li, Ziyi Wang, Yongliang Wu, Jianfei Cai, xu yang

Chain-of-Thought (CoT) prompting has emerged as a powerful approach to enhancing the reasoning capabilities of Large Language Models (LLMs). However, existing implementations, such as in-context learning and fine-tuning, remain costly and inefficient. To improve CoT reasoning at a lower cost, and inspired by the task vector paradigm, we introduce CoT Vectors, compact representations that encode task-general, multi-step reasoning knowledge. Through experiments with Extracted CoT Vectors, we observe pronounced layer-wise instability, manifesting as a U-shaped performance curve that reflects a systematic three-stage reasoning process in LLMs. To address this limitation, we propose Learnable CoT Vectors, optimized under a teacher–student framework to provide more stable and robust guidance. Extensive evaluations across diverse benchmarks and models demonstrate that CoT Vectors not only outperform existing baselines but also achieve performance comparable to parameter-efficient fine-tuning methods, while requiring fewer trainable parameters. Moreover, by treating CoT Vectors as a probe, we uncover how their effectiveness varies due to latent space structure, information density, acquisition mechanisms, and pre-training differences, offering new insights into the functional organization of multi-step reasoning in LLMs. The source code will be released.

Deep Learning · Theory

Vahid Shahverdi, Giovanni Luca Marchetti, Kathlén Kohn

We study function spaces parametrized by neural networks, referred to as neuromanifolds. Specifically, we focus on deep Multi-Layer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs) with an activation function that is a sufficiently generic polynomial. First, we address the identifiability problem, showing that, for almost all functions in the neuromanifold of an MLP, there exist only finitely many parameter choices yielding that function. For CNNs, the parametrization is generically one-to-one. As a consequence, we compute the dimension of the neuromanifold. Second, we describe singular points of neuromanifolds. We characterize singularities completely for CNNs, and partially for MLPs. In both cases, they arise from sparse subnetworks. For MLPs, we prove that these singularities often correspond to critical points of the mean-squared error loss, which does not hold for CNNs. This provides a geometric explanation of the sparsity bias of MLPs. All of our results leverage tools from algebraic geometry.

Wenjie Wei, Xiaolong Zhou, Malu Zhang, Ammar Belatreche, Qian Sun, Yimeng Shan, Dehao Zhang, Zijian Zhou, Zeyu Ma, Yang Yang 等

Spiking neural networks (SNNs) offer an energy-efficient alternative to traditional neural networks due to their event-driven computing paradigm. However, recent advancements in spiking transformers have focused on improving accuracy with large-scale architectures, which require significant computational resources and limit deployment on resource-constrained devices. In this paper, we propose a simple yet effective token pruning method for spiking transformers, termed TP-Spikformer, that reduces storage and computational overhead while maintaining competitive performance. Specifically, we first introduce a heuristic spatiotemporal information-retaining criterion that comprehensively evaluates tokens' importance, assigning higher scores to informative tokens for retention and lower scores to uninformative ones for pruning. Based on this criterion, we propose an information-retaining token pruning framework that employs a block-level early stopping strategy for uninformative tokens, instead of removing them outright. This also helps preserve more information during token pruning. We demonstrate the effectiveness, efficiency and scalability of TP-Spikformer through extensive experiments across diverse architectures, including Spikformer, QKFormer and Spike-driven Transformer V1 and V3, and a range of tasks such as image classification, object detection, semantic segmentation and event-based object tracking. Particularly, TP-Spikformer performs well in a training-free manner. These results reveal its potential as an efficient and practical solution for deploying SNNs in real-world applications with limited computational resources.

Applications · Time Series

Jeonghwan Im, Hyuk-Yoon Kwon

Deployed time-series forecasters suffer performance degradation under non-stationarity and distribution shifts. Test-time adaptation (TTA) for time-series forecasting differs from vision TTA because ground truth becomes observable shortly after prediction. Existing time-series TTA methods typically employ dual input/output adapters that indirectly modify data distributions, making their effect on the frozen model difficult to analyze. We introduce the Context-aware Output-Space Adapter (COSA), a minimal, plug-and-play adapter that directly corrects predictions of a frozen base model. COSA performs residual correction modulated by gating, utilizing the original prediction and a lightweight context vector that summarizes statistics from recently observed ground truth. At test time, only the adapter parameters (linear layer and gating) are updated under a leakage-free protocol, using observed ground truth with an adaptive learning rate schedule for faster adaptation. Across diverse scenarios, COSA demonstrates substantial performance gains versus baselines without TTA (13.91$\sim$17.03\%) and SOTA TTA methods (10.48$\sim$13.05\%), with particularly large improvements at long horizons, while adding a reasonable level of parameters and negligible computational overhead. The simplicity of COSA makes it architecture-agnostic and deployment-friendly. Source code: https://github.com/bigbases/COSA_ICLR2026

Deep Learning · Robustness

Junyi Chai, Shenyu Lu, Xiaoqian Wang

Fairness has become an arising concern in machine learning with its prevalence in decision-making processes, and the trade-offs between various fairness notions and between fairness and accuracy has been empirically observed. However, the inheritance of such trade-offs, as well as the quantification of the best achievable trade-offs, i.e., the Pareto optimal trade-offs, under varied constraints on fairness notions has been rarely and improperly discussed. Owing to the sub-optimality of fairness interventions, existing work fails to provide informative characterization regarding these trade-offs. In light of existing limitations, in this work, we propose a reformulation of the model-specific (MS) Pareto optimal trade-off, where we frame it as convex optimization problems involving fairness notions and accuracy w.r.t. the confusion vector. Our formulation provides an efficient approximation of the best achievable accuracy under dynamic fairness constraints, and yields systematical analysis regarding the fairness-accuracy trade-off. Going beyond the discussion on fairness-accuracy trade-offs, we extend the discussion to the trade-off between fairness notions, which characterizes the impact of accuracy on the compatibility between fairness notions. Inspired by our reformulation, we propose a last-layer retraining framework with group-dependent bias, and we prove theoretically the superiority of our method over existing baselines. Experimental results demonstrate the effectiveness of our method in achieving better fairness-accuracy trade-off, and that our MS Pareto frontiers sufficiently quantify the two trade-offs.

General Machine Learning · Transfer, Multitask and Meta-learning

Zichen Tian, Antoine Ledent, Qianru Sun

Scaling multi-task low-rank adaptation (LoRA) to a large number of tasks induces catastrophic performance degradation, such as an accuracy drop from 88.2% to 2.0% on DOTA when scaling from 5 to 15 tasks. This failure is due to parameter and representation misalignment. We find that existing solutions, like regularization and dynamic routing, fail at scale because they are constrained by a fundamental trade-off: strengthening regularization to reduce inter-task conflict inadvertently suppresses the essential feature discrimination required for effective routing. In this work, we identify two root causes for this trade-off. First, uniform regularization disrupts inter-task knowledge sharing: shared underlying knowledge concentrates in high-SV components (89% alignment on Flanv2→BBH). Uniform regularization forces high-SV components to update in orthogonal directions, directly disrupting the shared knowledge. Second, Conflict Amplification: Applying LoRA at the component-level (*e.g.*, $W_q, W_v$) amplifies gradient conflicts; we show block-level adaptation reduces this conflict with only 50% parameters. Based on these insights, we propose mtLoRA, a scalable solution with three novel designs: 1) Spectral-Aware Regularization to selectively orthogonalize low-SV components while preserving high-SV shared knowledge, 2) Block-Level Adaptation to mitigate conflict amplification and largely improve parameter efficiency, and 3) Fine-Grained Routing using dimension-specific weights for superior expressive power. On four large-scale (15-25 tasks) vision (DOTA and iNat2018) and NLP (Dolly-15k and BBH) benchmarks, mtLoRA achieves 91.7%, 81.5%, 44.5% and 38.5% accuracy on DOTA, iNat2018, Dolly-15k and BBH respectively, outperforming the state-of-the-art by 2.3% on average while using 47% fewer parameters and 24% less training time.

Applications · Chemistry and Drug Discovery

Zhe Zhang, Yuanning Feng, Yuxuan Song, Keyue Qiu, Hao Zhou, Wei-Ying Ma

AlphaFold3 introduces a diffusion-based architecture that elevates protein structure prediction to all-atom resolution with improved accuracy. This state-of-the-art performance has established AlphaFold3 as a foundation model for diverse generation and design tasks. However, its iterative design substantially increases inference time, limiting practical deployment in downstream settings such as virtual screening and protein design. We propose DCFold, a single-step generative model that attains AlphaFold3-level accuracy. Our Dual Consistency training framework, which incorporates a novel Temporal Geodesic Matching (TGM) scheduler, enables DCFold to achieve a 15× acceleration in inference while maintaining predictive fidelity. We validate its effectiveness across both structure prediction and binder design benchmarks.

Applications · Neuroscience, Cognitive Science

Yi-Chun Hung, Gregory Schwartz, Emily Cooper, Emma Alexander

Information processing in neural populations is inherently constrained by metabolic resource limits and noise properties, with dynamics that are not accurately described by existing mathematical models. Recent data, for example, shows that neurons in mouse visual cortex go into a "low power mode" in which they maintain firing rate homeostasis while expending less energy. This adaptation leads to increased neuronal noise and tuning curve flattening in response to metabolic stress. We have developed a theoretical population coding framework that captures this behavior using two novel, surprisingly simple constraints: an approximation of firing rate homeostasis and an energy limit tied to noise levels via biophysical simulation. A key feature of our contribution is an energy budget model directly connecting adenosine triphosphate (ATP) use in cells to a fully explainable mathematical framework that generalizes existing optimal population codes. Specifically, our simulation provides an energy-dependent dispersed Poisson noise model, based on the assumption that the cell will follow an optimal decay path to produce the least-noisy spike rate that is possible at a given cellular energy budget. Each state along this optimal path is associated with properties (resting potential and leak conductance) which can be measured in electrophysiology experiments and have been shown to change under prolonged caloric deprivation. We analytically derive the optimal coding strategy for neurons under varying energy budgets and coding goals, and show how our method uniquely captures how populations of tuning curves adapt while maintaining homeostasis, as has been observed empirically.

Applications · Robotics

Jinwoo Jang, Minjong Yoo, Sihyung Yoon, Honguk Woo

Language model (LM)-based embodied agents are increasingly deployed in real-world settings. Yet, their adaptability remains limited in dynamic environments, where constructing accurate and flexible world models is crucial for effective reasoning and decision-making. To address this challenge, we extend the Mixture-of-Experts (MoE) paradigm to embodied agents. While conventional MoE architectures modularize knowledge into expert components with pre-trained routing, they remain rigid once deployed, making them less effective for adapting to unseen domains in dynamic environments. We therefore propose Test-time Mixture of World Models (TMoW), a framework that enhances adaptability to unseen and evolving domains. TMoW updates its routing function over world models at test time, unlike conventional MoE where the function remains fixed, enabling agents to recombine existing models and integrate new ones for continual adaptation. It achieves this through (i) multi-granular prototype-based routing, which adapts mixtures across object- to scene-level similarities, (ii) test-time refinement that aligns unseen domain features with prototypes during inference, and (iii) distilled mixture-based augmentation, which efficiently constructs new models from few-shot data and existing prototypes. We evaluate TMoW on VirtualHome, ALFWorld, and RLBench benchmarks, demonstrating strong performance in both zero-shot adaptation and few-shot expansion scenarios, and showing that it enables embodied agents to operate effectively in dynamic environments.

Computer Vision · Image and Video Generation

Junzhe Li, Sifan Zhou, Liya Guo, Xuerui Qiu, Linrui Xu, TingTing Long, Chun Fan, Ming Li, Hehe Fan, Jun Liu 等

Unified multimodal models (UMMs) have emerged as a powerful paradigm in fundamental cross-modality research, demonstrating significant potential in both image understanding and generation. However, existing research in the face domain primarily faces two challenges: **(1) fragmentation development**, with existing methods failing to unify understanding and generation into a single one, hindering the way to artificial general intelligence. **(2) lack of fine-grained facial attributes**, which are crucial for high-fidelity applications. To handle those issues, we propose UniF$^2$ace, the first UMM specifically tailored for fine-grained face understanding and generation. **First**, we introduce a novel theoretical framework with a Dual Discrete Diffusion (D3Diff) loss, unifying masked generative models with discrete score matching diffusion and leading to a more precise approximation of the negative log-likelihood. Moreover, this D3Diff significantly enhances the model's ability to synthesize high-fidelity facial details aligned with text input. **Second**, we propose a multi-level grouped Mixture-of-Experts architecture, adaptively incorporating the semantic and identity facial embeddings to complement the attribute forgotten phenomenon in representation evolvement. **Finally**, to this end, we construct UniF$^2$aceD-1M, a large-scale dataset comprising *130K* fine-grained image-caption pairs and *1M* visual question-answering pairs, spanning a much wider range of facial attributes than existing datasets. Extensive experiments demonstrate that UniF$^2$ace outperforms existing models with a similar scale in both understanding and generation tasks, with 7.1% higher Desc-GPT and 6.6% higher VQA-score, respectively. Code is available in the supplementary materials.

Computer Vision · Vision Models & Multimodal

Zheng Liu, Mengjie Liu, Jingzhou Chen, Jingwei Xu, Bin CUI, Conghui He, Wentao Zhang

We introduce FLARE, a family of vision language models (VLMs) with a fully vision-language alignment and integration paradigm. Unlike existing approaches that rely on single MLP projectors for modality alignment and defer cross-modal interaction to LLM decoding, FLARE achieves deep, dynamic integration throughout the pipeline. Our key contributions include: (1) Text-Guided Vision Encoding that incorporates textual information during vision encoding to achieve pixel-level alignment; (2) Context-Aware Alignment Decoding that aggregates visual features conditioned on textual context during decoding for query-level integration; (3) Dual-Semantic Mapping Loss to supervise feature mapping from both modalities and enable modality-level bridging; and (4) Text-Driven VQA Synthesis that leverages high-quality text to generate VQA pairs and synthesize corresponding images, enabling data-level optimization. We train FLARE at 3B and 8B scales under both fixed and dynamic resolution settings, demonstrating that our full-modality alignment significantly outperforms existing methods while maintaining strong generalizability. FLARE 3B surpasses Cambrian-1 8B and Florence-VL 8B using only 630 vision tokens. Ablation studies reveal that FLARE achieves superior performance over existing methods with minimal computational cost. Even without dynamic resolution, FLARE outperforms LLaVA-NeXT, validating the effectiveness of our approach.

Applications · Language, Speech and Dialog

Tao He, Rongchuan Mu, Lizi Liao, Yixin Cao, Yang Li, Yijia Luo, Weixun Wang, Ming Liu, Bing Qin

Large reasoning models (LRMs) have shown strong performance in complex mathematical reasoning when optimized via reinforcement learning (RL). However, conventional outcome-only reward provides sparse feedback, leading to inefficient optimization. In this work, we investigate whether generative process reward models (GenPRMs) can accelerate RL training of LRMs by improving the utilization of reasoning trajectories. We first analyze critical limitations in existing GenPRMs, including their heavy reliance on reasoning ability during correctness judgment, and suppression of exploration as well as vulnerability to reward hacking during reward assignment. To address these limitations, we first propose a novel \textbf{intrinsic-signal-driven evaluation} mechanism, which judges reasoning steps using semantic cues from the solution, thus mitigating extensive dependence on GenPRM. Furthermore, we (i) adopt \textbf{thought-level rewarding granularity} to alleviate over-dense step rewards, and (ii) design a \textbf{difficulty-aware reward formulation} that dynamically balances exploration and exploitation and keeping the optimization target of key tokens to mitigate reward hacking. We integrate these innovations into the process reward-based GRPO, resulting in the proposed \textbf{TP-GRPO} algorithm. Experiments on LRMs with 1.5B and 7B parameters show that TP-GRPO achieves higher improvements while using significantly fewer training samples, and more analyses further confirm the effectiveness of our proposed process evaluation mechanism.