Zhiling Wang

Tianyang Han, Hengyu Shi, Junjie Hu et al.

Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal whether the reasoning trace is faithful, reliable, or useful to the model that consumes it. This outcome-only signal can reinforce traces that are right for the wrong reasons, overstate reasoning gains by rewarding shortcuts, and propagate flawed intermediate states in multi-step systems. To this end, we propose TraceLift, a planner-executor training framework that treats reasoning as a consumable intermediate artifact. During planner training, the planner emits tagged reasoning. A frozen executor turns this reasoning into the final artifact for verifier feedback, while an executor-grounded reward shapes the intermediate trace. This reward multiplies a rubric-based Reasoning Reward Model (RM) score by measured uplift on the same frozen executor, crediting traces that are both high-quality and useful. To make reasoning quality directly learnable, we introduce TRACELIFT-GROUPS, a rubric-annotated reason-only dataset built from math and code seed problems. Each example is a same-problem group containing a high-quality reference trace and multiple plausible flawed traces with localized perturbations that reduce reasoning quality or solution support while preserving task relevance. Extensive experiments on code and math benchmarks show that this executor-grounded reasoning reward improves the two-stage planner-executor system over execution-only training, suggesting that reasoning supervision should evaluate not only whether a trace looks good, but also whether it helps the model that consumes it.

Yifu Chen, Bingchen Huang, Zhiling Wang et al.

In-context learning (ICL) has become a classic approach for enabling LLMs to handle various tasks based on a few input-output examples. The effectiveness of ICL heavily relies on the quality of these examples, and previous works which focused on enhancing example retrieval capabilities have achieved impressive performances. However, two challenges remain in retrieving high-quality examples: (1) Difficulty in distinguishing cross-task data distributions, (2) Difficulty in making the fine-grained connection between retriever output and feedback from LLMs. In this paper, we propose a novel framework called TDR. TDR decouples the ICL examples from different tasks, which enables the retrieval module to retrieve examples specific to the target task within a multi-task dataset. Furthermore, TDR models fine-grained feedback from LLMs to supervise and guide the training of the retrieval module, which helps to retrieve high-quality examples. We conducted extensive experiments on a suite of 30 NLP tasks, the results demonstrate that TDR consistently improved results across all datasets and achieves state-of-the-art performance. Meanwhile, our approach is a plug-and-play method, which can be easily combined with various LLMs to improve example retrieval abilities for ICL. The code is available at https://github.com/Nnn-s/TDR.

Shuhang Tan, Hengyu Liu, Zhiling Wang

Relocalization is the basis of map-based localization algorithms. Camera and LiDAR map-based methods are pervasive since their robustness under different scenarios. Generally, mapping and localization using the same sensor have better accuracy since matching features between the same type of data is easier. However, due to the camera's lack of 3D information and the high cost of LiDAR, cross-media methods are developing, which combined live image data and Lidar map. Although matching features between different media is challenging, we believe cross-media is the tendency for AV relocalization since its low cost and accuracy can be comparable to the same-sensor-based methods. In this paper, we propose CMSG, a novel cross-media algorithm for AV relocalization tasks. Semantic features are utilized for better interpretation the correlation between point clouds and image features. What's more, abstracted semantic graph nodes are introduced, and a graph network architecture is integrated to better extract the similarity of semantic features. Validation experiments are conducted on the KITTI odometry dataset. Our results show that CMSG can have comparable or even better accuracy compared to current single-sensor-based methods at a speed of 25 FPS on NVIDIA 1080 Ti GPU.

Chao Jiang, Zhiling Wang, Shuhang Tan et al.

Kalman filters are widely used for object tracking, where process and measurement noise are usually considered accurately known and constant. However, the exact known and constant assumptions do not always hold in practice. For example, when lidar is used to track noncooperative targets, the measurement noise is different under different distances and weather conditions. In addition, the process noise changes with the object's motion state, especially when the tracking object is a pedestrian, and the process noise changes more frequently. This paper proposes a new estimation-calibration-correction closed-loop estimation method to estimate the Kalman filter process and measurement noise covariance matrices online. First, we decompose the noise covariance matrix into an element distribution matrix and noise intensity and improve the Sage filter to estimate the element distribution matrix. Second, we propose a calibration method to accurately diagnose the noise intensity deviation. We then propose a correct method to adaptively correct the noise intensity online. Third, under the assumption that the system is detectable, the unbiased and convergence of the proposed method is mathematically proven. Simulation results prove the effectiveness and reliability of the proposed method. Finally, we apply the proposed method to multiobject tracking of lidar and evaluate it on the official KITTI server. The proposed method on the KITTI pedestrian multiobject tracking leaderboard (http://www.cvlibs.net/datasets /kitti/eval_tracking.php) surpasses all existing methods using lidar, proving the feasibility of the method in practical applications. This work provides a new way to improve the performance of the Kalman filter and multiobject tracking.