8 papers
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
DeepSeek-AI, :, Anyi Xu +585
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation…
Continuous Limits of Coupled Flows in Representation Learning
Zilin Li, Weiwei Xu, Xuchun Tong +2
While modern representation learning relies heavily on global error signals, decentralized algorithms driven by local interactions offer a fundamental distributed alternative. Howe…
EmoLoom-2B: Fast Base-Model Screening for Emotion Classification and VAD with Lexicon-Weak Supervision and KV-Off Evaluation
Zilin Li, Weiwei Xu, Xuanbo Lu +1
We introduce EmoLoom-2B, a lightweight and reproducible pipeline that turns small language models under 2B parameters into fast screening candidates for joint emotion classificatio…
Neuro-Vesicles: Neuromodulation Should Be a Dynamical System, Not a Tensor Decoration
Zilin Li, Weiwei Xu, Vicki Kane
We introduce Neuro-Vesicles, a framework that augments conventional neural networks with a missing computational layer: a dynamical population of mobile, discrete vesicles that liv…
DuoTeach: Dual Role Self-Teaching for Coarse-to-Fine Decision Coordination in Vision--Language Models
Wei Yang, Yiran Zhu, Zilin Li +3
Coarse-to-fine path decision-making requires predicting a valid taxonomy path in which earlier decisions constrain later ones. However, existing benchmarks score each level indepen…
Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents
Zihao Wang, Xujing Li, Yining Ye +24
We present Game-TARS, a generalist game agent trained with a unified, scalable action space anchored to human-aligned native keyboard-mouse inputs. Unlike API- or GUI-based approac…
NeuroGaze-Distill: Brain-informed Distillation and Depression-Inspired Geometric Priors for Robust Facial Emotion Recognition
Zilin Li, Weiwei Xu, Xuanqi Zhao +1
Facial emotion recognition (FER) models trained only on pixels often fail to generalize across datasets because facial appearance is an indirect and biased proxy for underlying aff…
Augment the Pairs: Semantics-Preserving Image-Caption Pair Augmentation for Grounding-Based Vision and Language Models
Jingru Yi, Burak Uzkent, Oana Ignat +4
Grounding-based vision and language models have been successfully applied to low-level vision tasks, aiming to precisely locate objects referred in captions. The effectiveness of g…