9 papers
S3Mem: Structured Spatiotemporal Scene-Event Memory for Long-Horizon Interactive Question Answering
Encheng Su, Jianyu Wu, Jinouwen Zhang +9
Long-horizon memory question answering often requires sparse evidence from heterogeneous histories, including events, object states, visual observations, temporal relations, and ca…
CPRet: A Dataset, Benchmark, and Model for Retrieval in Competitive Programming
Han Deng, Yuan Meng, Shixiang Tang +2
Competitive programming benchmarks are widely used in scenarios such as programming contests and large language model assessments. However, the growing presence of duplicate or hig…
SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration
Ye Li, Yuan Meng, Zewen Sun +7
Vision-Language-Action (VLA) models have attracted increasing attention for their strong control capabilities. However, their high computational cost and low execution frequency hi…
EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance
Yingxin Li, Ye Li, Yuan Meng +4
As large language models (LLMs) continue to advance, the demand for higher quality and faster processing of long contexts across various applications is growing. KV cache is widely…
Q-DiT: Accurate Post-Training Quantization for Diffusion Transformers
Lei Chen, Yuan Meng, Chen Tang +5
Recent advancements in diffusion models, particularly the architectural transformation from UNet-based models to Diffusion Transformers (DiTs), significantly improve the quality an…
PRANCE: Joint Token-Optimization and Structural Channel-Pruning for Adaptive ViT Inference
Ye Li, Chen Tang, Yuan Meng +5
We introduce PRANCE, a Vision Transformer compression framework that jointly optimizes the activated channels and reduces tokens, based on the characteristics of inputs. Specifical…