Linlin Yang
Publications
Contextual Information Policy Optimization for Search Agents
Search agents extend large language models beyond static parametric memory by enabling them to acquire and use external evidence during multi-step reasoning. For knowledge-intensive tasks involving complex or evolving information, their reliability depends not only on retrieving relevant evidence but also on using it to guide subsequent reasoning. However, existing methods primarily reward final-answer correctness or intermediate progress, without directly assessing whether post-retrieval actions are grounded in the retrieved evidence. This misalignment encourages prior-driven reasoning: agents form conclusions based on internal knowledge and use retrieval mainly to confirm them, resulting in confirmation bias and inefficient evidence use. To address this issue, we propose Contextual Information Policy Optimization (CIPO), an evidence-oriented reinforcement learning framework that explicitly aligns policy optimization with external evidence use. CIPO assigns dense, turn-level credit to reasoning actions influenced by retrieved information, while combining this evidence-use signal with a global outcome reward to preserve answer correctness. With this manner, CIPO discourages evidence-detached guesses and promotes reasoning trajectories in which retrieved facts can guide or revise subsequent reasoning. Importantly, CIPO requires neither human process annotations nor an additional reward model. Extensive experiments on seven in-domain and out-of-domain benchmarks show that CIPO reduces the prevalence of prior-driven reasoning and achieves excellent performance on most tasks.
Kalman Meets Curriculum: Efficient Dynamic Prompt Selection for Adaptive RL Finetuning
Reinforcement learning (RL) finetuning significantly enhances the reasoning capabilities of large language models (LLMs), yet its effectiveness critically depends on selecting prompts of appropriate difficulty for the current policy. This is challenging because prompt difficulty evolves throughout training. Existing online methods therefore face a trade-off: evaluation-based approaches are accurate but expensive, while prediction-based approaches are efficient but typically assume stationary difficulty, making them ill-suited to RL's non-stationary training dynamics. To address these issues, we propose a Kalman-Guided Prompt Selection method (KGPS), which reformulates prompt selection as a dynamic state estimation problem rather than static difficulty prediction. KGPS models each prompt's latent success rate in logit space using a linear-Gaussian state-space model, with process noise coupled to the magnitude of policy updates so that uncertainty increases when the policy changes more substantially. A Kalman filter then maintains a calibrated Gaussian posterior over prompt difficulty, and prompts are selected by maximizing a posterior-expected training utility that favors intermediate-difficulty prompts while naturally revisiting uncertain ones. The resulting procedure is adaptive to policy drift and requires no additional rollouts beyond standard policy training. Extensive experiments across mathematics, planning, and geometry reasoning benchmarks, as well as multiple RL algorithms, show that KGPS consistently improves both final accuracy and rollout efficiency over strong baselines, establishing state-of-the-art performance among online prompt selection methods. For example, on DeepSeek-R1-Distill-7B, KGPS uses 83% fewer rollouts than DS while even improving the average performance by 0.12 point across six math reasoning benchmarks.
LDFE: Laplacian Decoupled Feature Enhancement Block for Dual-Stream CNN-based RGB-IR Object Detection
The complementary information between RGB and IR images can significantly enhance object detection performance under extreme conditions. Existing methods prefer dual-stream CNN backbones built upon YOLO for feature extraction and focus on the design of feature fusion. In this paper, we introduce the Laplacian Decoupled Feature Enhancement block (LDFE) to fuse features from different stages of the dual-stream CNN backbone. By design, LDFE simultaneously considers the characteristics of modalities and structures for feature fusion by employing global-local decomposition, denoising, fusion, and reconstruction, sequentially. The LDFE first separates features into global and local components based on Laplacian Pyramid, and then performs denoising and fusion based on Global State Space Enhancement module (GS2E) and Local Convolutional Correlation Enhancement module (LC2E) separately. Specifically, the GS2E conducts a two-branch architecture for the main and auxiliary modalities. It dynamically suppresses noise in the main modality through cross-modal attention derived from the auxiliary modality, while employing a State Space Model to capture long-range dependencies within the global feature representations of the main modality. To obtain bidirectional interaction, the two modalities systematically alternate their main/auxiliary roles. Moreover, the LC2E suppresses noise in local features and leverages spatial and channel dimension along with triple convolution to extract fine-grained details for fusion. These innovative designs achieve a significant performance improvement, with mAP surpassing the SOTA methods 6.2%, 3.7%, 4.7%, 2.3%, 4.1% and 2.0% on M3FD, DroneVehicle, LLVIP, FLIR-Aligned, KAIST and VEDAI datasets,respectively.
Kirchhoff-Inspired Neural Networks for Evolving High-Order Perception
Deep learning architectures are fundamentally inspired by neuroscience, particularly the structure of the brain's sensory pathways, and have achieved remarkable success in learning informative data representations. Although these architectures mimic the communication mechanisms of biological neurons, their strategies for information encoding and transmission are fundamentally distinct. Biological systems depend on dynamic fluctuations in membrane potential; by contrast, conventional deep networks optimize weights and biases by adjusting the strengths of inter-neural connections, lacking a systematic mechanism to jointly characterize the interplay among signal intensity, coupling structure, and state evolution. To tackle this limitation, we propose the Kirchhoff-Inspired Neural Network (KINN), a state-variable-based network architecture constructed based on Kirchhoff's current law. KINN derives numerically stable state updates from fundamental ordinary differential equations, enabling the explicit decoupling and encoding of higher-order evolutionary components within a single layer while preserving physical consistency, interpretability, and end-to-end trainability. Extensive experiments on partial differential equation (PDE) solving and ImageNet image classification validate that KINN outperforms state-of-the-art existing methods.
Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization
Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs due to its extensive group-based sampling requirement. While recent selective data utilization methods can mitigate this overhead, they could induce estimation bias by altering the underlying sampling distribution, compromising theoretical rigor and convergence behavior. To address this limitation, we propose Dynamic Pruning Policy Optimization (DPPO), a framework that enables dynamic pruning while preserving unbiased gradient estimation through importance sampling-based correction. By incorporating mathematically derived rescaling factors, DPPO significantly accelerates GRPO training without altering the optimization objective of the full-batch baseline. Furthermore, to mitigate the data sparsity induced by pruning, we introduce Dense Prompt Packing, a window-based greedy strategy that maximizes valid token density and hardware utilization. Extensive experiments demonstrate that DPPO consistently accelerates training across diverse models and benchmarks. For instance, on Qwen3-4B trained on MATH, DPPO achieves 2.37$\times$ training speedup and outperforms GRPO by 3.36% in average accuracy across six mathematical reasoning benchmarks.