Hanwang Zhang lists you as their PhD student. Claim this profile to confirm it and keep the rest of your record right.
Claim this profileAcademic lineage
View as a treeAdvisors
Works30 from public data
- Distilling Causal Effect of Data in Class-Incremental Learning218
The proposed causal effect distillation method can improve various state-of-the-art CIL methods by a large margin and capture the Incremental Momentum Effect of the data stream, which can help to retain the old effect overwhelmed by the new data effect, and thus alleviate the forgetting of the old class in testing.
- 164
- Risk Assessment Model for UAV Cost-Effective Path Planning in Urban Environments130
The results show that the risk-cost-based path planning method can generate safer path for UAV operations than the traditional shortest-distance-based method.
- A new consensus theory-based method for formation control and obstacle avoidance of UAVs124
Simulation results demonstrate that the ICA can make UAVs with different initial states form the specified formation while satisfying all the constraints, and a safe and efficient flight for UAV formation can be ensured with the obstacle avoidance algorithm.
- 118
- Learning to Segment the Tail96
This work proposes a “divide&conquer” strategy for the challenging LVIS task: divide the whole data into balanced parts and then apply incremental learning to conquer each one, which derives a novel learning paradigm: class-incremental few-shot learning, which is especially effective for the challenge evolving over time.
- Number it: Temporal Grounding Videos like Flipping Manga86
Number-Prompt (NumPro) is introduced, a novel method that empowers Vid-LLMs to bridge visual comprehension with temporal grounding by adding unique numerical identifiers to each video frame by transforming VTG into an intuitive process: flipping through manga panels in sequence.
- Multi-constrained cooperative path planning of multiple drones for persistent surveillance in urban environments47
The multi-constrained cooperative path planning (MCCPP) algorithm is developed to maximize the combinational coverage area of drones while giving preference to the area that hasn’t been visited beyond a certain time interval.
- Multiple air route crossing waypoints optimization via artificial potential field method45
An Artificial Potential Field (APF) model is proposed considering attractive forces produced by the optimal routes and repulsive forces generated by obstacles to optimize the different airspace topologies varying the number of CWPs, air route segments and PRDs.
- Trajectory-based flight scheduling for AirMetro in urban environments by conflict resolution28
AirMetro is a new concept proposed for future 3D public air transportation service, which carries the passengers commuting from a point to the destination, routinely and periodically, and the trajectory-based flight scheduling method can improve the flight efficiency by ensuring the flight safety of UAVs and reducing the average delay of Uavs as well in AirMetro.
- 23
- 22
- Mimic In-Context Learning for Multimodal Tasks17
MimIC more strictly approximates the shift effects of ICDs by integrating lightweight learnable modules into LMMs with four key enhancements: inserting shift vectors after attention layers, assigning a shift vector to each attention head, and employing a layer-wise alignment loss.
- 10
- 7
- 6
- 6
- 4
- 4
- 3
- LayerCake: Token-Aware Contrastive Decoding within Large Language Model Layers3
This work introduces a token-aware, layer-localized contrastive decoding method that aligns specific token types with their most influential transformer layers to improve factual generation and consistently improves factuality across multiple LLMs and various benchmarks.
- Hierarchical progressive learning for zero-shot peptide-HLA binding prediction and automated antigenic peptide design3
An automated peptide mutation search program is developed, which automatically identifies binding peptides for any target HLA class-I allele and mutates weak-binding or non-binding peptides to enhance binding affinity under the guidance of HPL models.
- 3
- 3
- 2
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-11. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it. You will be asked to sign in.